ArticlePDF Available

Taking care of business: the impact of culture and gender on entrepreneurs’ blended value creation goals

Authors:

Abstract and Figures

We examine entrepreneurs’ economic, social, and environmental goals for value creation for their new ventures. Drawing on ethics of care and theories of societal post-materialism, we develop a set of hypotheses predicting patterns of value creation across gender and countries. Using a sample of 15,141 entrepreneurs in 48 countries from the Global Entrepreneurship Monitor, we find that gender and cultural values of post-materialism significantly impact the kinds of value creation emphasized by entrepreneurs. Specifically, women entrepreneurs are more likely than men to emphasize social value goals over economic value creation goals. Individuals who start ventures in strong post-materialist societies are more likely to have social and environmental value creation goals and less likely to have economic value creation goals. Furthermore, as levels of post-materialism rise among societies, the relationship between value creation goals and gender changes, intensifying both the negative effect of being female on economic value goals and the positive effect on social value goals. In other words, post-materialism further widens the gender gap in value creation goals.
This content is subject to copyright. Terms and conditions apply.
1 23
Small Business Economics
An Entrepreneurship Journal
ISSN 0921-898X
Small Bus Econ
DOI 10.1007/s11187-016-9747-4
Taking care of business: the impact of
culture and gender on entrepreneurs’
blended value creation goals
Diana M.Hechavarría, Siri A.Terjesen,
Amy E.Ingram, Maija Renko, Rachida
Justo & Amanda Elam
1 23
Your article is protected by copyright and all
rights are held exclusively by Springer Science
+Business Media New York. This e-offprint is
for personal use only and shall not be self-
archived in electronic repositories. If you wish
to self-archive your article, please use the
accepted manuscript version for posting on
your own website. You may further deposit
the accepted manuscript version in any
repository, provided it is only made publicly
available 12 months after official publication
or later and provided acknowledgement is
given to the original source of publication
and a link is inserted to the published article
on Springer's website. The link must be
accompanied by the following text: "The final
publication is available at link.springer.com”.
Taking care of business: the impact of culture and gender
on entrepreneurs’ blended value creation goals
Diana M. Hechavarrı
´a.Siri A. Terjesen .
Amy E. Ingram .Maija Renko .
Rachida Justo .Amanda Elam
Accepted: 24 May 2016
Springer Science+Business Media New York 2016
Abstract We examine entrepreneurs’ economic,
social, and environmental goals for value creation
for their new ventures. Drawing on ethics of care and
theories of societal post-materialism, we develop a set
of hypotheses predicting patterns of value creation
across gender and countries. Using a sample of 15,141
entrepreneurs in 48 countries from the Global
Entrepreneurship Monitor, we find that gender and
cultural values of post-materialism significantly
impact the kinds of value creation emphasized by
entrepreneurs. Specifically, women entrepreneurs are
more likely than men to emphasize social value goals
over economic value creation goals. Individuals who
start ventures in strong post-materialist societies are
more likely to have social and environmental value
creation goals and less likely to have economic value
creation goals. Furthermore, as levels of post-materi-
alism rise among societies, the relationship between
value creation goals and gender changes, intensifying
both the negative effect of being female on economic
D. M. Hechavarrı
´a
Center for Entrepreneurship, College of Business,
University of South Florida, 4202 E. Fowler Ave., Tampa,
FL 33620, USA
e-mail: dianah@usf.edu
S. A. Terjesen (&)
Management Department, American University,
Washington, DC 20008, USA
e-mail: terjesen@american.edu
S. A. Terjesen
Norwegian School of Economics, Bergen, Norway
A. E. Ingram
College of Business and Behavioral Sciences, Clemson
University, 139 Sirrine Hall, Clemson, SC 29634, USA
e-mail: amyi@clemson.edu
M. Renko
Managerial Studies, University of Illinois at Chicago, MC
243, University Hall 2211, 601 South Morgan Street,
Chicago, IL 60607, USA
e-mail: maija@uic.edu
R. Justo
IE Business School, Maria de Molina 11, 28006 Madrid,
Spain
e-mail: rachida.justo@ie.edu
A. Elam
Center for Women’s Entrepreneurial Leadership, Babson
College, Wellesley, MA, USA
e-mail: aelam@babson.edu
A. Elam
Management, Innovation, and Entrepreneurship
Department, Poole College of Management, North
Carolina State University, Raleigh, NC 27695, USA
123
Small Bus Econ
DOI 10.1007/s11187-016-9747-4
Author's personal copy
value goals and the positive effect on social value
goals. In other words, post-materialism further widens
the gender gap in value creation goals.
Keywords Environmental entrepreneurship
Female entrepreneurship Global Entrepreneurship
Monitor Gender Social entrepreneurship Value
creation Venture goals
1 Introduction
Although entrepreneurs are generally assumed to
pursue economic gains, research suggests that foun-
ders pursue a diverse range of value creation targets
(Emerson 2003; Stephan et al. 2016). Reflecting this
reality, entrepreneurship research has expanded
beyond a focus on economic profits to include value
creation goals of societal progress and environmental
preservation, including social and environmental
value creation (Cohen et al. 2008; Hall et al. 2010;
Cohen and Winn 2007; York et al. 2016). Building on
this philosophy, our research takes a comparative
approach to identify the antecedents that shape
entrepreneurs’ ‘‘blended value’’ goals (Zahra et al.
2014) across societies and genders.
The blended value framework, sometimes referred
to as the ‘‘triple bottom line’’ (Elkington 2004),
describes how organizations create value on multiple
levels, particularly economic, social, and environmen-
tal value creation targets (Cohen et al. 2008). Despite
increased awareness and rhetoric of blended value
(Dembek et al. 2015; Pavlovich and Corner 2014),
there is no systematic assessment of what predicts
blended value goals. This is surprising since pursuing
blended value enables organizations to capture effi-
ciencies not possible through pursuing purely com-
mercial, purely social, or purely environmental
strategies (Bonini and Emerson 2005). Extant research
focuses on the antecedents of economic, social, or
environmental goals separately, e.g., entrepreneurs’
self-interested motivations for economic gains, per-
sonal growth, and autonomy (Gatewood 1993; Shaver
et al. 2001). Individuals’ prosocial values drive them
to start ventures (Agafonow 2014; Miller et al. 2012;
Santos 2012). As entrepreneurs often pursue several
goals simultaneously (Williams and Nadin 2011) with
varying emphasis on each goal (Gorgievski et al.
2011), we seek to identify multi-level drivers of
blended value.
Recent studies examine the factors that are associated
with entrepreneurial value creation goals at either
individual (Gorgievski et al. 2011), organizational, or
institutional (Cohen and Winn 2007; Stephan et al. 2014)
levels. Individual psychological factors such as self-
efficacy or compassion are associated with entrepreneurs’
social goals (Miller et al. 2012; Smith and Woodworth
2012). Formal and informal national institutions such as
government regulations and cultural norms are associated
with individual engagement in commercial ventures
(Terjesen et al. 2016), social enterprises (Stephan et al.
2014), and environmental businesses (Vazquez-Brust
and Sarkis 2012). However, this burgeoning research
fails to investigate the joint, multi-level effects of these
factors (Miller et al. 2012). Our research question is: How
do gender and national cultural values shape entrepre-
neurs’ goal to create ventures that deliver economic,
social, or environmental value to the market?
We link the moral theory of ethics of care and justice
(Gilligan 1982) with post-materialism to examine
individual- and institutional-level drivers of entrepre-
neurs’ blended value creation goals. Ethics of justice is a
masculine-oriented value system prioritizing fairness,
rights, and obligations. Ethics of care, a feminine-
oriented value system, focuses on the interconnected-
ness among parties involved, and nurturing (Flanagan
and Jackson 1987). We draw on research that finds
women are more likely than men to express an ethic of
care (Jaffee and Hyde 2000) to argue that women may
also be potentially more likely to place greater priorities
on social and environmental value creation goals over
economic goals. Considering that enterprising individ-
uals are embedded in contexts (Welter and Smallbone
2010;Hughesetal.2012), we investigate how blended
value creation goals are shaped by a national culture of
post-materialism—the extent to which a national soci-
ety emphasizes autonomy, self-expression, and well-
being in shaping venture value creation objectives. We
argue that entrepreneurs in strong post-materialist
societies will prioritize social and environmental value
creation goals over economic goals.
We test our model (Fig. 1) with multi-level multi-
variate regression. We find that women and men
nascent entrepreneurs and business owners [here-
inafter entrepreneurs], on average, exhibit significant
heterogeneity in the blended value goals of their
ventures, depending on their cultural context.
D. M. Hechavarrı
´a et al.
123
Author's personal copy
Specifically, compared to men, women entrepreneurs
in post-materialist societies tend to report significantly
lower economic value creation goals and significantly
higher social value creation goals. Therefore, although
globally there appear to be distinct gender differences
in entrepreneurs’ value creation goals, these differ-
ences are attenuated or intensified in the cultural
context. Thus, our findings demonstrate that culture is a
powerful construct that impacts female entrepreneurs
differently, accounting for previously unexplained
variance in value creation goals across economies.
Our study makes several contributions. First, we
provide evidence of the multidimensional nature of
entrepreneurs’ blended value goals. Second, we describe
the individual and national factors associated with
entrepreneurial goals, extending recent literature on cul-
tural influences on social and environmental value creation
(Stephan et al. 2014; Vazquez-Brust and Sarkis 2012).
Third, we answer calls to address the connection between
venture value creation goals and gender (Jennings and
Brush 2013). Although prior research suggests a link
between gender and value creation goals (Hechavarria
et al. 2012), the present study provides more compelling
evidence through multi-level multivariate regression to
concurrently test variation in blended value within and
among societies. Fourth, by focusing on how national
culture shapes between-gender differences in venture
value creation motives, we respond to calls from gender
scholars to examine the heterogeneity among women and
men entrepreneurs (Hughes and Jennings 2012; Hughes
et al. 2012;deBruinetal.2006; Terjesen et al. 2015).
2 Venture value creation goals: blended value
creation framework
Organizations’ decisions and actions reflect the
founders’ values and the social conditions in which
Gender:
Female
Controls:
Age
Household Income
Education
Equal Income
Good Career
Status
Media
Established Business
Necessity
Number of Owners
Innovativeness
Male-dominated Industry
Extractive industry
Transformative industry
Business Serv. industry
Entrepreneurs’ Value
Creation Goals:
Economic / Social /
Environmental
Cultural context:
Level of Post-
materialism
Controls:
GDP (per capita)
% GDP growth
% Unemployment
% Labor Force Partic.
% Tertiary Education
Ecological Footprint
% TEA Necessity (2008)
Country level
Individual level
H2a, H2b, H2c
H1a, H1b, H1c
Fig. 1 Proposed study
model of gender and
culture’s impact on the
blended value creation of
organizations
Taking care of business: the impact of culture and gender on entrepreneurs’ blended
123
Author's personal copy
they were created (Stinchcombe 1965). Values are
desirable outcomes or modes of conduct which guide
behavior (Rokeach 1973) and are inherently subjec-
tive because they frame individuals’ judgments. Value
creation is a process through which business activities
create outputs that capture value for the organization,
its stakeholders, society, and the environment (Bow-
man and Ambrosini 2000; Elkington 2004), and is
immensely subjective (Lepak et al. 2007). Individuals
create value through a wide range of interactions,
activities, and relationships in their organizations
(Lepak et al. 2007) which are socially constructed
entities (Campbell 2000). Value creation goals are
qualitative or quantitative targets that guide strategic
action. We focus on founders’ value creation goals as
early goal choices can substantially influence the types
and culture of organizations created (Pettigrew 1979).
For example, an organization’s social and/or environ-
mental responsibility position is often based on the
values of powerful individual champions (Bansal and
Roth 2000).
To advance our understanding of the multidimen-
sional nature of value creation goals, we draw on the
blended value creation framework. The blended value
creation framework is a normative ethical lens to
examine how organizations create, add, or destroy
economic, social, and environmental value (Bugg-
Levine and Emerson 2011). The framework suggests
that economic, social, and environmental value cre-
ation goals are compatible (Cohen et al. 2008; Hall
et al. 2010) and that founders must decide how to
combine and prioritize each objective as part of an
overall value proposition. Pursuing blended value
enables entrepreneurs to capture efficiencies not
possible through pursuing purely commercial, purely
social, or purely ecological business strategies. Con-
sequently, blended value creation frameworks reflect a
founder’s desire to balance and maximize multiple
value sets.
3 Ethics of care, ethics of justice, and blended
value creation goals
Ethics of care is a decision-making framework that
focuses on the needs of others (Gilligan 1982;
Flanagan and Jackson 1993; Tronto 1993), and the
corresponding responsibilities and relationships. Ethic
of care is associated with values of involvement,
empathy, sustaining harmonious relationships,
extended communicative rationality, and holism. In
contrast, ethics of justice is an impartial decision-
making framework grounded in ethical universalism
to safeguard impartial and equal treatment of all
individuals (Kohlberg 1981; Rawls 2001), and con-
cerns rights and rules. Ethics of justice is associated
with autonomy, objectivity, positivistic rationality,
reductionism, and universality. After three decades of
exploration in philosophy and feminist research, the
ethics of care lens is now used to interpret manage-
ment phenomena (e.g., Lawrence and Maitlis 2012;
Bampton and Maclagan 2009).
Although ‘‘care’’ is implied in the blended value
creation framework, most business ethics and corpo-
rate social responsibility texts either ignore ethic of
care or offer only a cursory mention (Hawk 2011).
Gilligan (1982) maintains that there are fundamental
differences between women and men’s framing of
moral issues in such a way that women prioritize
communal, relational values (e.g., Schwartz and Rubel
2005) which are the ‘‘guiding principles’’ associated
with ethical guardianship (Noddings 2012). Women
also define power in terms of taking care of others
(Hawk 2011). Thus, women’s care ethic leads them to
foster values of empathy, sympathy, compassion,
loyalty, discernment, love, benevolence, community,
and promotion of a civil society more readily than men
(Held 2005). These two moral orientations are not
gender-specific but rather gender-related as both
genders use the perspectives with scholars arguing
that ‘‘women are most unlikely to take only the justice
perspective, as some men are claimed to, at least until
some mid-life crisis jolts them into ‘‘bifocal’’ moral
vision’’ (Baier 1985, p. 56). Therefore, it should be no
surprise that some comparative empirical studies show
that women have higher care orientations than do men
(Calhoun 2012; Jaffee and Hyde 2000; Luthar and
Karri 2005). Despite a well-established stream of
gender scholarship in the general entrepreneurship
literature, there are only a few descriptive reports of
social entrepreneurship and gender (Bosma and Levie
2010; Harding and Cowling 2006; Hechavarria et al.
2012; Lepoutre et al. 2013). These studies indicate that
men are more likely to start social businesses than are
women, and that the gap between male and female
participation in entrepreneurial activity is larger for
commercial entrepreneurship than for social
entrepreneurship (Estrin et al. 2013; Lepoutre et al.
D. M. Hechavarrı
´a et al.
123
Author's personal copy
2013), suggesting women’s care orientations may
drive social value goals.
Ethical framing also has implications for a special
kind of helping: pro-environmental behavior (Blocker
and Eckberg 1997). As noted by Zelezny et al. (2000),
environmentalism reflects an extended ‘‘other’’ orien-
tation where women are more likely than men to be
concerned about environmental issues due to an
extension of an ethic of care and a corresponding
desire to protect nature (Hunter et al. 2004). As a
result, women who pursue ecological value creation
are generally motivated by ethical concerns, while
men are motivated more by the promise of a better
competitive advantage or operational savings for their
businesses (Braun 2010).
In sum, the gender socialization literature suggests
a link between ethics of care and variations in blended
value creation targets between men and women. Men
prioritize money and career, reflecting a more promi-
nent ethics of justice. Conversely, women are more
focused on social provisioning by maintaining rela-
tionships, helping others, and helping nature (Power
2008; Luthar and Karri 2005). Consequently, we
hypothesize these factors will affect entrepreneurs’
blended value goals:
H1a Compared to men, women entrepreneurs are
less likely to report economic value creation goals.
H1b Compared to men, women entrepreneurs are
more likely to report social value creation goals.
H1c Compared to men, women entrepreneurs are
more likely to report environmental value creation
goals.
3.1 Culture: post-materialism and blended value
creation goals
A rich literature in comparative international
entrepreneurship indicates that national-level institu-
tions are associated with entrepreneurial activity (Ter-
jesen et al. 2016). Sociologists often argue that group-
level, societal, and cultural constructs drive individuals’
preferences (Erbring and Young 1979). For example,
American business leaders are more likely to utilize
‘ethics of justice’’ while their Turkish counterparts
resort to ‘‘ethics of care’’ (Simga-Mugan et al. 2005).
Entrepreneurs in some countries may have, on average,
higher levels of economic, social, or environmental
value creation goals than entrepreneurs in other coun-
tries due to systematic differences in national culture
(Gatignon et al. 1989)—the set of normative values and
beliefs regarding what is considered desirable and
undesirable in a society (Javidan and House 2001). As
another example, entrepreneurs who live in cultures that
support wealth creation and autonomy are morelikely to
reflect these values. Conversely, a cultural context that
supports altruistic and ecological awareness champions
entrepreneurs who pursue social and environmental
value creation targets. Culture can also affect con-
sumers’ needs (Roth 1995), thus impacting how
founders aim to satisfy such needs.
We contend that culture, understood in terms of
post-materialism, influences the compositions of
blended value creation goals. According to Inglehart
(1981,2000,2003), post-materialism is the process of
transforming from a more materialistic culture with an
emphasis on economic and physical security to a post-
materialistic society with a greater emphasis on life
goals such as autonomy and self-expression. As a
result, materialist cultures prioritize achieving finan-
cial success, acquiring substantial possessions, pro-
jecting the right image, and attaining high status and
substantial worldly goods (Belk 1984; Inglehart 1981;
Richins and Dawson 1992). In contrast, post-materi-
alist societies prioritize humanism, quality of life,
peace, human rights, the environment, love, esteem,
and self-actualization (Knutsen 1990). The shift
toward post-materialism coincides with the transition
from modernized societies (those societies that have
already experienced industrialization) to post-modern
societies. Post-modern societies rely more on intel-
lectual capabilities than on physical inputs/natural
resources, as evidenced by prominent service and
technology sectors (Hechavarria 2015). Accordingly,
as countries become wealthier, they tend to move
toward post-materialism, and their populations
become increasingly interested in the well-being of
others and the environment (Inglehart 2000), civic
activism (Inglehart 2003), ecological consideration,
and pro-environmental attitudes (Pakulski et al. 1998),
even if these values conflict with economic growth
(Banarjee and McKeage 1994). As a result, when
levels of post-materialism increase in a given society,
citizens become less self-focused and become more
aware of their interdependence with nature and with
the social world (Wilson 2005).
Taking care of business: the impact of culture and gender on entrepreneurs’ blended
123
Author's personal copy
The relationship between post-materialism and
entrepreneurship is explored in a series of papers by
Uhlaner et al. (2002) and Uhlaner and Thurik
(2004,2007). Overall, they report a negative relation-
ship between post-materialism and entrepreneurial
activity: countries marked by higher levels of post-
materialism have lower total (both nascent and new
business) commercial entrepreneurial activity rates.
However, other studies find that level of post-materi-
alism is positively associated with individuals’
engagement in social entrepreneurship (Stephan
et al. 2014; Hechavarria 2015), and overall society
level of engagement of social entrepreneurship.
Therefore, as post-materialism transforms societies
from a materialistic to a less materialistic focus,
entrepreneurs’ goals shift from prioritizing financial
profits to pursuing ecological and social value cre-
ation. Accordingly, we argue that post-materialism
affects entrepreneurs’ blended value creation goals:
H2a Entrepreneurs in post-materialistic countries
are less likely to report economic value creation than
those in materialistic countries.
H2b Entrepreneurs in post-materialistic countries
are more likely to report social value creation than
those in materialistic countries.
H2c Entrepreneurs in post-materialistic countries
are more likely to report environmental value creation
than those in materialistic countries.
3.2 The interplay of gender and culture on blended
value creation
Cultural environments generally promote individual
dispositions and social relations that are congruent
with the environment and that inhibit non-conforming
behaviors (Schwartz 2006). Since cultural norms and
ideals impact ethical reasoning (Husted and Allen
2008), this suggests that these beliefs attenuate or
strengthen ethic of care or justice values between male
and female founders (Held 2005).
Within society, broadly held cultural beliefs shape
the social and economic standards that entrepreneurs
must adhere to in order to successfully operate their
businesses (Smith et al. 2002). The distinction
between weak and strong post-materialist value
orientations across countries is therefore relevant to
changing standards of acceptability with regard to care
and justice value sets. Accordingly, there is potential
prejudice when the dominant societal beliefs promote
a stereotype about women that is incompatible with
the perceived characteristics required for success as a
business founder. Thus, embeddedness in a post-
materialist context may facilitate the internalization
and expression of non-economic organizational values
among entrepreneurs, particularly women.
Women’s entrepreneurial activities may be more
strongly affected by some cultural forces. For example,
compared to men, women are less likely to engage in
entrepreneurship in countries with hostile institutional
environments (Estrin and Mickiewicz 2011). Overall,
cultural contexts more strongly affect women’s entre-
preneurial intentions and perceptions as compared to
men’s (Bullough et al. 2014; Elam 2008;Santosetal.
2014). If women’s businesses are more susceptible to
the surrounding societies’ values and expectations, post-
materialism may moderate the relationship between
gender and value creation intentions.
By supporting human development and well-being
(Tronto 1993), ethics of care orientations overlap
strongly with the values championed by post-materi-
alism. Therefore, humanism and quality of life
promoted in post-materialist societies might reinforce
the positive effect on altruistic venturing goals among
women entrepreneurs, further increasing women
founders’ likelihood of expressing social and envi-
ronmental values. Gender is generally a background
identity such that gender’s effects on behavior and
evaluations vary by context; however, we argue that it
is important to systematically identify gender differ-
ences across contexts. Indeed, gender’s effect on
certain behaviors depends on the salience of gender to
the situation (Ridgeway and Smith-Lovin 1999).
Specifically, some attitudes described by Inglehart’s
(1981) as post-materialist are linked to feminist ideals
(Campbell 2004), indicating correlations among post-
materialism, feminism, and gender. For example,
advanced industrial democracies’ shift toward post-
materialistic values is associated with moves toward
feminist ideals. This, in turn, can facilitate gender
equality and egalitarianism among societies that
maintain higher levels of post-materialism (Hayes
et al. 2000), providing an environment that encourages
female founders to more freely express inherent values
linked toward their care orientations. As a result, we
contend that the societal emphasis of post-materialism
D. M. Hechavarrı
´a et al.
123
Author's personal copy
intensifies the negative relationship between female
gender and economic value creation, as well as the
positive relationship between female gender and
social and environmental value creation goals:
H3a Post-materialism moderates the relationship
between women entrepreneurs and economic value cre-
ation goals such that this negative relationship is stronger in
countries with higher levels of post-materialism.
H3b Post-materialism moderates the relationship
between women entrepreneurs and social value creation
goals such that this positive relationship is stronger in
countries with higher levels of post-materialism.
H3c Post-materialism moderates the relationship
between women entrepreneurs and environmental
value creation goals such that this positive relationship
is stronger in countries with higher levels of post-
materialism.
4 Data and methods
We draw data from four sources: the global
entrepreneurship monitor (GEM) (Reynolds et al.
2005), World Value Surveys (WVS)/European Values
Survey (EVS) (Inglehart and Welzel 2005), the Happy
Planet Index (HPI), and the World Bank’s World
Development Indicators (WB-WDI) (2009).
We utilize GEM data from 48 countries in 2009 (the
only year that GEM included questions on blended
value goals, our dependent variable) to capture
individual data on value creation goals.
1
GEM mea-
sures the drivers and consequences of entrepreneurial
activity globally (Bergmann et al. 2014) and is used in
over 1000 publications and policy documents. The
GEM database offers harmonized, comparable data on
many types of entrepreneurship, including early-stage
‘nascent’ efforts in dozens of countries. GEM data
facilitates evidence-based policies that guide
entrepreneurship (see Reynolds et al. 2005; Bosma
2013) and can be used with other data sources
(Davidsson 2006) as GEM includes a small list of
single item measures each year. GEM is extremely
useful for multi-level research designs with a micro-
level dependent variable and micro- and macro-
independent variables (Bergmann et al. 2014; Bosma
2013). To prevent ecological fallacy, other data should
only be added at the aggregate macro-level. See
Table 1for all variables.
We combine macro-level data from three databases
with the 2009 GEM data. WVS/EVS conducts face-to-
face surveys of nationally representative samples of
respondents of more than 90 societies over six waves
(1981–2009). Each participating country has at least
1000 sampled respondents in each wave. HPI exam-
ines 151 societies’ sustainable well-being in terms of
life expectancy, experienced well-being, and ecolog-
ical footprint. WB-WDI compiles regional, national,
and international data.
Our sample includes all GEM 2009 entrepreneur
respondents with complete information on the inde-
pendent and dependent variables of interest
(n=15,141).
2
We analyze the venturing goals of
three populations grouped together as entrepreneurs:
nascent entrepreneurs, baby business owners, and
established business owners.
3
The sample is weighted
to each country’s census adult labor force.
We utilize a multi-level multivariate regression to
jointly analyze the impact of gender and culture.
Multivariate regression models are simultaneous
regression systems that provide joint estimates from
several regression models, each with its own
1
Countries: Algeria, Argentina, Belgium, Bosnia and Herze-
govina, Brazil, Chile, China, Colombia, Croatia, Denmark,
Dominican Republic, Finland, France, Germany, Greece,
Guatemala, Hong Kong, Hungary, Iceland, Iran, Israel, Italy,
Japan, Jordan, Latvia, Lebanon, Malaysia, Morocco, Nether-
lands, Norway, Peru, Romania, Russia, Saudi Arabia, Serbia,
Slovenia, South Africa, South Korea, Spain, Switzerland,
Tunisia, Uganda, UK, US, Uruguay, Venezuela, West Bank
and Gaza Strip, and Yemen.
2
We apply Rubin’s (1987) multiple imputation procedure to
estimate the missing data values for the controls: household
income education, industry, and innovativeness using the
logistic regression method, and linear regression method for
age (using predictors: GDP, gender, education, age, household
income, and industry, and innovativeness). The percent of cases
imputed from the covariates are all under one-third: household
income (21 %), education (2 %), industry (12 %), and innova-
tion (27 %). After five versions of the data set were generated by
the multiple imputation procedure, we analyzed each version of
imputed data. The goal is to estimate a model separately in each
dataset, and identify the best-fitting model.
3
GEM methods and sampling frame are reported in Reynolds
et al. (2005). Nascent entrepreneurs are actively involved in the
creation or development of a venture in which they will have
ownership do not yet have positive cash flows. Baby business
owners have ownership share in a new venture that has positive
cash flow and is less than 42 months old. Established business
owners’ ventures are operational and older than 42 months.
Taking care of business: the impact of culture and gender on entrepreneurs’ blended
123
Author's personal copy
Table 1 Variables’ coding, source, and prior research
Variable Coding Source Prior research
Individual level
Social, economic, or
environmental value creation
Ratio: Three ipsative measures based on response to: ‘‘Organizations may
have goals according to the ability to generate economic value, societal
value, and environmental value’’ and then ‘‘Please allocate a total of 100
points across these three categories as it pertains to your goals’’
GEM APS Blended value framework (Elkington 1994; Cohen
et al. 2008; Lepoutre et al. 2013; Bacq et al. 2013)
Age Continuous: entrepreneurs’ age GEM APS Age influences entrepreneurial activity (Gartner
et al. 2004), social entrepreneurship intentions
(Nga and Shamuganathan 2010), and
environmental attitudes (Zelezny et al. 2000)
Household income Ordinal: lowest third, middle third, and upper third GEM APS Founder’s available resources may affect the
organization’s resources (Gartner et al. 2004)
Education Ordinal: five categories that capture: no educational background, some
secondary education, secondary education, post-secondary education,
and graduate experience
GEM APS Education influences propensity toward and type of
entrepreneurial activity (Bosma 2013; Estrin et al.
2016; Pathak and Muralidharan 2016)
Established Businesses Binary: 1 =venture or owner/managers is a business older than
42 months old, and 0 =the venture or owner/managers is a business
younger than 42 months and is nascent or early-stage venturing
GEM APS Organizational age might impact strategies
(Sternberg and Wennekers 2005)
Number of Owners Continuous: total number of venture’s owners GEM APS The number of owners influences venture attitudes
and goals (Cliff 1998)
Innovative Orientation Ordinal: very high innovative orientation =9; very low/no innovative
orientation =1 (Bosma et al. 2012) This additive three-item index based
on responses to: potential customers will consider this product or service
new or unfamiliar (1 =none, 1 =some, 3 =many); how many
businesses offer the same product (1 =none, 2 =some, 3 =many);
and whether technologies or procedures required for this product or
service been available for less than a year, or between 1 and 5 years, or
longer than 5 years (1 =over five, 2 =between five to one; 3 =less
than one). Scores vary between 1 and 3 if respondents indicate that they
‘do not know’’ to any combination of the questions
GEM APS Innovativeness is more closely linked to economic
values than to social values (Desa and Kotha
2006; Ahuja and Lampert 2001; Shane and Stuart
2002)
Necessity motivation Binary: ‘‘Are you involved in this start-up to take advantage of a business
opportunity or because you have no better choices for work?’’ If the
founder indicated that he/she had no better options for work, or that he/
she was dissatisfied with his/her job and that there were better options for
work =1, all other responses =0
GEM APS Individuals’ contextual motivations are reflected in
the characteristics and goals of their organizations
(Bosma 2013)
Male-dominated Industry Binary: Men comprise at least 75 % of the labor force. Male-dominated
industries: agriculture, fishing, hunting, forestry, mining, manufacturing,
utilities, transportation, construction, communications, electric, gas, and
sanitary services, and wholesale trade. Non-male-dominated industries
include real estate, insurance, and finance retail trade, public
administration, education, and health services
GEM APS Male-dominated industries can influence social
norms around gender roles (Anna et al. 2000)
D. M. Hechavarrı
´a et al.
123
Author's personal copy
Table 1 continued
Variable Coding Source Prior research
Industry Nominal: captures four categories of industries based on GEM
classification system of (1) extractive, (2) transforming, (3) business
services, (4) consumer oriented
GEM APS Certain industries may be more aligned with
blended value goals than others (Cohen et al.
2008)
Social Perceptions: Equal Income Binary: All inhabitants prefer uniform living standard. 1 =yes; 0 =no GEM APS Social perceptions of what is acceptable in society
can impact the kind of venturing activity
(Reynolds 2010)
Social Perceptions: Career Binary: Starting a business is considered as a good career choice. 1 =yes;
0=no
GEM APS
Social Perceptions: Status Binary: Persons growing a successful new business receive high status.
1=yes; 0 =no
GEM APS
Social Perceptions: Media Binary: Lots of media coverage for new businesses. 1 =yes; 0 =no GEM APS
Female Binary: 0 =male, 1 =female GEM APS Gender impacts business goals (Estrin et al. 2013;
Zelezny et al. 2000)
Country level
Economic development level Continuous: Gross domestic product (per capita) WB GDP is positively related to social entrepreneurship
(Hoogendoorn and Hartog 2011)
% GDP Growth Ratio: % change in GDP PPP from 2008 to 2009 WB Higher rates of entrepreneurship are associated with
growing economies (Reynolds 2010)
% Tertiary education Ratio: % of population with tertiary education WB Knowledge helps individuals to identify
opportunities, shapes value orientations (Busenitz
et al. 2000), and is related to post-materialist
values (Moor 2003)
% Labor force part. Ratio: % of population active in the labor force WB Greater shares of the population participating in the
labor force population are accompanied by
stronger focus on economic value creation
(Verheul et al. 2006)
% Unemployment Ratio: % of population actively unemployed WB
Gender Stereotypes Interval: Measures attitudes toward gender stereotypes in employment,
income, political power and education (Cronbach a=0.85): ‘‘When
jobs are scarce, men should have more right to a job than women;
Problem if women have more income than husband; On the whole, men
make better political leaders than women do; A university education is
more important for a boy than for a girl.’’ All items are on a four-point
scale where 1 indicates ‘‘strongly agree’’ and 4 is ‘‘strongly disagree.’
Therefore, a low score indicates a strong climate gender stereotypes,
whereas a high score indicates a low climate of gender stereotypes. Data
are then aggregated to compute an average representative score to
estimate the degree of gender stereotypical values held by societies
analyzed in our study
WVS Gender stereotypes push women into career choices
to which they earn less than their husbands
(Tinsley et al. 2015)
Taking care of business: the impact of culture and gender on entrepreneurs’ blended
123
Author's personal copy
Table 1 continued
Variable Coding Source Prior research
Ecological Footprint Ratio: Per capita measure of the amount of land required to sustain a
country’s consumption patterns, measured in terms of global hectares,
which represent a hectare of land with average productive bio-capacity
HPI Ecological footprint is related to environmental
degradation (Hall et al. 2010), leading to indirect
entrepreneurial opportunities
% TEA Necessity (2008) The percentage of the adult population engaged in necessity
entrepreneurship for the prior year (2008)
GEM According to institutional logics, the kind of
venturing activity prevalent in a society can
perpetuate the kinds of ventures pursued
(Hechavarria and Reynolds 2009)
Level of Post-materialism Ratio: Based on the following question and responses: ‘‘There is a lot of
talk these days about what this country’s goals should be in the next 10
or 15 years. Would you please say which one of them you yourself
consider most important in the long-run: (a) Maintaining the order of
nation; (b) Giving the people more say in important government
decisions; (c) Fighting rising prices; or (d) Protecting freedom of
speech.’’ Respondents rank the two most important items. Regardless of
the order, items a and c correspond with materialist values; items b and d
correspond with post-materialist values (Inglehart 1977). Based on the
responses in these questions, the respondents are given a score from one
to three in the 4-item index post-materialism index. The first and third
options tap materialist priorities, while the second and fourth options tap
post-materialist priorities. If both materialist items are given high
priority, the score =1; if both post-materialist items are given high
priority, the score =3; if one materialist item and one post-materialism
item are given high priority the score =2. We follow Tranter and
Western (2008) in recoding the 4-item post-materialism index based on
the aggregate frequencies for post-materialism using all WVS/EVS
waves of data weighted by population for percentage of population that
reported 3 ‘‘pure post-materialist’’ values
WVS Prior research uses country-level aggregate values
rather than individual values, of the Inglehart
index. We follow prior work in interpreting the
index as a measure of shared (rather than
personal) attitudes for each country (Hansen and
Tol 2003). The 4-item index was selected. Post-
materialism significantly impacts both social and
commercial entrepreneurship (Stephan et al. 2014;
Uhlaner and Thurik 2007)
GEM APS Global Entrepreneurship Monitor Adult Population Survey, GEM NES Global Entrepreneurship Monitor National Expert Survey, WB World Bank, WVS World Values
Survey, HPI Happy Planet Index
D. M. Hechavarrı
´a et al.
123
Author's personal copy
contemporaneously correlated error term. This
method estimates all equations’ parameters simulta-
neously with a generalized least squares estimator
which takes the covariance structure of the residuals
into account. Multi-level multivariate regression con-
tains both fixed and random effects. Fixed effects are
directly estimated, in addition to being indirectly
estimated by covariances of random intercepts and
slopes. The multi-level multivariate regression system
allows us to examine a nested structure, and lets
coefficients vary from one context to another (Chib
and Greenberg 1995). We apply multivariate regres-
sion techniques to consider Iblended value goals
through the joint estimation of a system of equations
(Zellner 1962):
yik ¼b0iþx1ikb1iþuik ;
i¼1;2;...;I;k¼1;2;...;K;
where y
ik
is the blended value indicator for ith blended
value goal for the kth entrepreneur, b
0i
a coefficient,
x
1ik
a19q
i
vector of q
i
regressors specific to blended
value goal i,b
1i
aq
i
91 vector of coefficients, and u
ik
is an error with E(u
ik
)=0. Stacking the kentrepre-
neurs yields the model for Igoals:
yi¼b0iþX1ib1iþui;i¼1;2;...;I;
or
y1
y2
...
yI
2
6
6
6
4
3
7
7
7
5
¼
b01
b02
...
bI
2
6
6
6
4
3
7
7
7
5
þ
X11 0... 0
0X12 ... 0
...
00... X1I
2
6
6
6
4
3
7
7
7
5
b11
b12
...
b1I
2
6
6
6
4
3
7
7
7
5
þ
u1
u2
...
uI
2
6
6
4
3
7
7
5
where y
i
,b
0i
and u
i
are k91 vectors, X
1i
ak9q
i
matrix, and b
1i
is a q
i
91 vector. If a nascent or early-
stage entrepreneur’s score kon two goals iand pis
related by unobservable factors, then u
ik
is correlated
with u
pk
for i=p. Multivariate models allow for such
correlation:
Eu
iku0
ph

¼rip;if k¼hand 0 otherwise
where kand hdenote two different individuals. We
extend the multi-level framework to consider multiple
dependent variables simply by recognizing that
blended value goals are clustered, in our case within
countries (Hauck and Street 2006). By considering
individual entrepreneurs’ blended value goals as the
data hierarchy’s lowest tier, we investigate the possi-
bility of within-individual and within-country corre-
lation of blended value goals. Our two-level multi-
level model involves the set of Iblended values goals
(level 1) clustered within Jcountries (level 2):
Yij ¼b0iþx1ijb1iþu0ij þe0ij ;
i¼1;2;...;I;j¼1;2;...;J:
Thus, y
ij
is the ith blended value goal clustered within
the jth country. We assume that error terms u
0
iand e
0
ij
are normally distributed with zero mean and constant
variance (r
u,i
,r
e,i
,) for each blended value goal.
Country-level error e
0
ij captures the random error for
blended value goals iin the jth country. We consider
the possibility that blended value goals are correlated
within countries with cov(e
0
ij,e
0
pj)=re,ip.
Consistent with multi-level models, the intra-class
correlation coefficient is estimated as:
ICCMVML ¼s2
u0ij s2
u0ij þs2
e0ij

1
;0\ICCMVML\1:
We use Stata’s GSEM command (Rabe-Hesketh
et al. 2004).
4.1 Dependent variables
Our dependent variable captures blended value cre-
ation goals in terms of: economic value goals,social
value goals, and environmental value goals, and is
derived from GEM’s adult population survey (APS).
Individuals were read the statement ‘‘Organizations
may have goals according to the ability to generate
economic value, societal value, and environmental
value’’ and then asked to ‘‘Please allocate a total of
100 points across these three categories as it pertains to
your [venture’s] goals.’’ We compute three continuous
measures for economic, social, and environmental
value goals according to their responses, thus
acknowledging that entrepreneurs can have multiple
value creation goals. The dependent variable is
ipsative, or forced choice, as respondents’ three
potential value creation goals must sum 100 %. For
instance, a respondent could allocate 30 % to eco-
nomic goals, 60 % to social goals, and 10 % to
environmental goals. If an entrepreneur values all
Taking care of business: the impact of culture and gender on entrepreneurs’ blended
123
Author's personal copy
three components as highly important, he/she would
assign 33.33 % to economic goals, 33.33 % to social
goals, and 33.33 % to environmental goals.
The usefulness of ipsative measures is a matter of
debate (Cornwall and Dunlap 1994). The major
advantages of ipsative measurement are that respon-
dents are forced to make comparisons and that this
choice scaling is on the same dimension with the same
meaning. Ipsative measures make intuitive sense to a
subject because they mimic the practical situation
where one must decide among alternative approaches
(Baron 1996). The major disadvantage of ipsative
measures is that the answers depend on one other, thus
variables’ intercorrelations are nearly all negative.
However, multivariate regressions with correlated
error terms help mitigate the problems associated
with the ipsative dependent variable intercorrelations
because the ‘‘multivariate’’ errors are contemporane-
ously correlated in our model.
4.2 Independent variables
4.2.1 Post-materialism
Post-materialism comes from the WVS/EVS (Ingle-
hart 1977,2003) surveys’ five multiyear waves since
1981 (1981–1984; 1989–1993; 1994–1999;
1999–2004; 2005–2009),
4
totaling some 80 countries
(Gorodnichenko and Roland 2011). We aggregate the
weighted pooled percentage of the population indi-
cating post-materialism values using the four-item
index (item Y002) which is the most complete data
across countries; this choice is consistent with Uhlaner
and Thurik (2007). Inglehart’s (1977, 2008) post-
materialist index has long been incorporated into
Eurobarometer (since 1970) and WVS (since the early
1980s) and is based on the following question and four
responses: ‘‘There is a lot of talk these days about what
this country’s goals should be in the next 10 or
15 years. Would you please say which one of them
you yourself consider most important in the long-run:
(a) Maintaining the order of nation; (b) Giving the
people more say in important government decisions;
(c) Fighting rising prices; or (d) Protecting freedom of
speech.’’ Respondents rank the two most important
items. Materialist values reflect physical or economic
insecurity; in these societies, respondents choose
items a and c. Respondents in countries with post-
materialist values choose b and d (Inglehart 1977).
When both materialist items are high priority, the post-
materialism score =1; both post-materialist items
high priority =3; and one materialist item and one
post-materialism item high priority =2.
We follow prior research (e.g., Abramson and
Inglehart 1995; Gelissen 2007; Inglehart and Abram-
son 1994; Davis and Davenport 1999) in using
Inglehart’s country-level rather than individual-level
values. We interpret the index as a measure of shared
rather than personal attitudes for each country in our
sample (Hansen and Tol 2003). We calculate the
aggregate percent of respondents in each country
holding ‘‘pure’’ post-materialist (=3) values using all
WVS/EVS waves.
5
This population-weighted pooled
percent of respondents indicates the average level of
post-materialist values for waves 1–6, for each
country. This measure ranges from 0 to 100.
4.2.2 Sex
Respondents’ self-reported sex (from GEM) is coded
one for female, and zero for male.
4.3 Control variables
Table 1summarizes each control variable’s rationale
based on prior research (see Bosma 2013 for a
summary of all GEM-based studies). We include 22
control variables from GEM, WB, and HPI: age,
household income, education, established business,
number of owners, industry, male-dominated industry,
innovativeness, necessity motivation, career percep-
tions of entrepreneurship, equal income perceptions,
status perceptions of entrepreneurship, media percep-
tions of entrepreneurship, GDP per capita (PPP),
percent GDP growth, percent unemployment, percent
of the population with a tertiary education, percent of
4
For Tunisia, Hong Kong, Lebanon, Yemen, and West Bank
and Gaza, data was not available for Y002 in waves 1–5, so we
used wave 6 responses to estimate societal averages (Wave 6
was administered from 2010 to 2014).
5
We pool all available waves of WVS/EVS to ensure the
largest proportion of coverage on the post-materialism index
since not every country in our sample participated in every wave
of the WVS/EVS. This approach is acceptable and common in
prior research using post-materialism (e.g., Abramson et al.
1997) since work by Inglehart and Baker (2000) finds cultural
values are stable constructs over time.
D. M. Hechavarrı
´a et al.
123
Author's personal copy
the population that is actively participating in the labor
force, gender stereotypes index, ecological footprint,
and percentage of adult population engaged in neces-
sity entrepreneurship.
4.4 Sample descriptive statistics
Table 2provides descriptive and bivariate statistics for
all variables. The respondents’ mean age is almost
40 years. About 37 % of the sample is female. On
average, 15.3 % are in the lowest third percentile for
household income, 37.4 % in the middle third, and
47.2 % in the upper third. Approximately 11 % of
respondents have no educational background, 20 %
have some secondary education, 35 % have secondary
education, 32 % have post-secondary education, and
2 % have graduate education. About 47 % of sample
owns a business that is older than 3.5 years. The
average size is about two owners. The average level of
innovation across entrepreneurial ventures is moderate
at 5.2 on a scale of 1–9. Of all venturing activity, about
4 % is in male-dominated industries, globally. Like-
wise, about 20 % are in transforming, 11 % in
extractive, 11 % in business services, and 58 % in
consumer-oriented industries. Approximately 62 % of
entrepreneurs believe that people should have equal
income and approximately 72 % of entrepreneurs
consider that entrepreneurship is a means to attain
higher status in society. Roughly 61 % of entrepre-
neurs believe that the media portrays entrepreneurship
positively. The average GDP (per capita) for our
sample of countries is $20,073, with a -0.38 % GDP
growth rate in 2009. The average unemployment rate is
9.75 %, with 61 % active labor force. The average
level of tertiary education is 51 % for countries.
Gender stereotypical views about men’s and women’s
roles are moderate at 2.32 of 4. The ecological footprint
average is 3.56 global hectares; the societal average of
post-materialist values is 12 %. On average, 3.7 % of
adults engage in necessity entrepreneurship.
5 Results
Table 2highlights some interesting features of the
bivariate analysis. We find preliminary evidence that
gender impacts entrepreneurs’ triple bottom line value
goals such that being female is positively correlated
with social goals and negatively correlated with
economic goals. However, there is no significant
correlation with environmental value goals and being
female. Furthermore, post-materialist culture influ-
ences triple bottom line value orientations such that
post-materialism is positively correlated to both social
and environmental goals, and negatively correlated
with economic goals. A few country-level variables
are significant and strongly correlated, such as
ecological footprint and GDP (r=0.891; p\0.01)
and ecological footprint and percent tertiary enroll-
ment (r=0.752; p=\0.01). To confirm that there
are no multicollinearity issues in our subsequent
analysis, we calculate variance inflation factors
(VIFs). Mean VIF is 1.95, with a high 5.16 and a
low of 1.01 (well below the threshold of 10).
We run three simultaneous multi-level multivariate
regression systems. We first calculate the null (inter-
cept only) model to obtain the interclass correlation
coefficients (ICCs) or between country variation
(Rabe-Hesketh and Skrondal 2006). Second, we
estimate random intercept and fixed slope models
with predictors. Third, we add cross-level interactions.
Table 3presents the null model results to identify
variation in blended value creation goals associated
with inter-individual differences (among respondents
in different countries) and intra-individual differences
(among individuals within countries). Table 3’s ICC
indicates that about 9 % of economic, 23 % of social,
and 34 % of environmental value goal variations are
due to between country differences. ICCs under 0.05
suggest that a one-level model with robust estimation
of standard errors is appropriate (Aguinis et al. 2013).
In our case, there is sufficient evidence to pursue
multi-level modeling using a multivariate regression
system.
Next, we estimate random intercept and fixed slope
models for blended value creation goals including our
controls (see Table 4). Subsequently, we estimate
random intercept and fixed slope models for blended
value creation goals including covariates and predic-
tors of our main effects (see Table 5). This model
predicts economic, social, and environmental values
based on a common intercept, control variables, and
our predictors: gender and post-materialism. Table 5
shows a direct single-level effect of gender on
economic value (b=-
2.098; p=\0.0001) and
social value (b=1.892; p=\0.0001). However, we
find no evidence of a direct effect of gender on
environmental value (b=0.159; p=0.340). In other
Taking care of business: the impact of culture and gender on entrepreneurs’ blended
123
Author's personal copy
Table 2 Descriptive statistics and bivariate correlations
Variable
Mean
S.D.
1
2
3
4
5
6
7
8
9
10
11
12
1. Economic Value Goals
65.38
26.36
1
2. Social Value Goals
21.63
19.51
-.827**
1
4. Age
39.98
11.75
-.036**
.013
.046**
1
5. Household Income
2.36
0.77
-.010
.007
.008
.020**
1
6. Education
1.76
1.28
-.119**
.101**
.076**
-.005
.281**
1
7. Established Business Owner
0.47
0.50
.047**
-.039**
-.032**
.316**
.038**
-.045**
1
8. Owners
1.95
6.99
-.033**
.016*
.037**
-.044**
.006
.019**
-.067**
1
9. Innovative Orientation
5.16
1.16
-.028**
.019**
.025**
-.084**
.042**
.043**
-.047**
.004
1
10. Necessity Motivation
0.36
0.48
.058**
-.061**
-.022**
-.020**
-.142**
-.183**
-.016*
-.009
-.003
1
11. Male-dominated industry
0.04
0.21
-.005
-.019**
.035**
-.046**
-.028**
-.020**
-.183**
.014
-.007
.025**
1
12. Transforming Industry
0.20
0.40
.026**
-.040**
.008
.049**
.045**
.010
.036**
.001
.015*
-.018*
.143**
1
14. Business Services
0.11
0.32
-.005
.012
-.006
.006
.073**
.180**
-.009
.012
.028**
-.064**
-.074**
-.177**
15. Consumer Oriented
0.58
0.03
-.013
.005
.015*
-.002
.005
.010
-.024**
-.002
.008
-.006
.116**
.040**
16. Equal Income
0.62
0.49
-.013
-.014
.041**
-.006
-.048**
-.031**
-.029**
-.003
-.001
.028**
.042**
-.009
17. Good Career Choice
0.68
0.47
.052**
-.056**
-.017*
-.093**
-.048**
-.152**
-.053**
-.024**
.025**
.085**
.048**
-.009
18 New Business Status
0.72
0.45
.047**
-.022**
-.054**
-.070**
-.015*
-.066**
-.044**
-.019*
.006
.032**
.030**
-.034**
19. New Business Media
0.61
0.49
-.001
.003
-.001
-.048**
-.033**
-.083**
-.060**
-.018*
.020**
.064**
.052**
-.019*
20. GDP per capita (PPP)
20073.99
14005.00
-.198**
.145**
.159**
.275**
.119**
.348**
.150**
-.012
-.036**
-.239**
-.085**
.011
21. % GDP Growth
-0.38
5.67
.159**
-.094**
-.157**
-.199**
-.138**
-.333**
-.086**
-.022**
.035**
.177**
.039**
-.047**
22. % Unemployment
9.75
5.20
.070**
-.088**
-.008
-.027**
.090**
.037**
.004
.001
.047**
.011
.026**
.040**
23. % Labor Force
60.61
7.80
-.043**
.023**
.045**
.044**
.025**
-.063**
.050**
-.066**
-.008
.020**
-.061**
-.010
24. % Tertiary Enrollment
50.92
23.85
-.143**
.063**
.170**
.225**
.155**
.319**
.113**
-.012
.013
-.154**
-.072**
.072**
25. Gender Stereotypes
2.32
0.30
-.090**
.027**
.125**
.053**
.018*
-.017*
.027**
-.022**
-.011
.003
-.025**
.043**
26. Ecological Footprint
3.56
1.67
-.200**
.139**
.169**
.261**
.112**
.337**
.143**
-.016*
-.019**
-.231**
-.077**
.034**
27. % TEA Necessity (2008)
3.69
3.86
0.142**
-0.102**
-0.117**
-0.214**
-0.207**
-0.355**
0.018*
0.207**
0.176**
0.245**
-0.118**
0.239**
28. Female
0.37
0.48
-.046**
.054**
.008
-.026**
-.113**
-.043**
-.052**
.000
-.034**
.040**
-.005
-.069**
29. Post-materialism
12.24
6.83
-.190**
.152**
.135**
.184**
.077**
.209**
.063**
-.004
-.024**
-.182**
-.076**
-.020**
Variable
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
14. Business Services
-.124**
1
15. Consumer Oriented
-.009
-.009
1
16. Equal Income
.005
-.023**
-.002
1
17. Good Career Choice
.055**
-.062**
-.010
.113**
18 New Business Status
.034**
-.025**
-.013
.083**
.226**
1
19. New Business Media
.051**
-.043**
-.003
.070**
.191**
.209**
1
20. GDP per capita (PPP)
-.125**
.144**
.016*
-.052**
-.242**
-.167**
-.198**
1
23. % Labor Force
-.047**
-.043**
.000
-.028**
-.007
-.001
.005
.030**
.138**
-.210**
1
24. % Tertiary Enrollment
-.167**
.134**
.014
-.018*
-.186**
-.180**
-.242**
.694**
-.633**
.143**
.102**
1
25. Gender Stereotypes
-.016*
.017*
.015*
.037**
-.006
-.076**
-.035**
.063**
-.130**
.087**
.271**
.222**
1
26. Ecological Footprint
-.114**
.157**
.018*
-.057**
-.233**
-.167**
-.197**
.891**
-.685**
-.084**
.080**
.752**
.079**
1
27. % TEA Necessity (2008)
-0.016**
-0.015**
0.073**
0.201**
-0.049**
-0.123**
0.112**
-0.726**
0.667**
-0.286**
0.110**
-0.739**
-0.006**
0.112**
1
28. Female
-.003
-.052**
-.019**
.025**
-.002
.020*
.019*
-.037**
.036**
-.065**
.036**
-.052**
.053**
-.039**
-0.691
1
29. Post-materialism
-.136**
.064**
.005
.012
-.142**
-.046**
-.097**
.584**
-.412**
-.250**
.186**
.397**
.330**
.605**
.016*
-0.363
*** p \0.01; ** p\0.05; * p\0.01
D. M. Hechavarrı
´a et al.
123
Author's personal copy
words, being female decreases the likelihood of
reporting economic value creation goals by about
2 % while increasing the likelihood of reporting social
value creation goal by about 1.9 %.
We also find a direct cross-level effect of post-
materialism on economic value (b=-0.299;
p=0.055), social value (b=0.193; p=0.027),
and environmental value (b=0.105; p=0.044).
For a one percent increase in post-materialism, there
is a 0.299 % decrease in the relative importance of
economic value goals. Conversely, for every 1 %
increase in post-materialism, there is a 0.193 %
increase in the relative importance of social value
goals, and a 0.105 % increase in the relative impor-
tance of environmental goals. The average level of
post-materialism in our sample is 12 %; the highest
level is 27 %. So a society’s transition from a
moderate level of post-materialism to the highest
level of post-materialism (*15 % increase in the
society’s post-materialism) translates into a 4.5 %
decrease in the relative importance of economic goals,
a 2.9 % increase in the relative importance of social
goals, and a 1.6 % increase in the relative importance
of environmental goals reported by entrepreneurs.
Subsequently, we estimate a random intercept and
fixed slope model for economic, social, and environ-
mental value scores based on intercept, controls,
gender, post-materialism, and an interaction term for
gender and post-materialism to estimate the cross-
level interaction effect on the dependent variables (see
Table 6). We find that women entrepreneurs in post-
materialist societies are significantly less likely to
emphasize economic goals (b=-0.206; p=0.001)
and significantly more likely to emphasize social goals
(b=0.216; p\0.0.001). However, the data do not
indicate a cross-level interaction between gender and
post-materialism for environmental goals (b=
-0.021; p=0.560). Findings in Table 6suggest that
the difference in slope between women’s and men’s
economic goals is about -0.206, and 0.216 for social
goals. The slope for women in post-materialist soci-
eties is -0.435 (-0.229 -0.206 =-0.435) for eco-
nomic and 0.626 (0.118 ?0.216 =0.626) for social
goals. A society transitioning from a moderate to the
highest level of post-materialism (*15 % increase in
the society’s post-materialism) translates into an
additional 9.39 % increase in the relative importance
of social value goals among women, and an additional
6.52 % decrease in the relative importance of eco-
nomic goals among women.
We also correlate the predicted values from the last
multivariate regression estimated model with the
cross-level interaction to the actual reported scores
among respondents to obtain a pseudo R
2
. According
to this estimate, our full model (with cross-level
interaction) explains about 24 % of the variance for
economic value, 26 % for social value, and 24 % for
environmental value goals.
Figure 2a presents the post-estimated values based
on our cross-level interaction models for economic
value goals. The results indicate that male and female
entrepreneurs are less likely to prioritize economic
goals in post-materialist countries, with a stronger
effect for female entrepreneurs leading to an increase
in the size of the gender gap. Male entrepreneurs in
societies with post-materialism levels at one standard
Table 3 Simultaneous multi-level multivariate regression null model
Level and variable Economic Social Environmental
Estimate SE pvalue Estimate SE pvalue Estimate SE pvalue
Level 1
Intercept 64.783 7,687,266.000 0.000 21.988 499,753.000 0.000 13.190 0.659 0.000
Variance components
Within-Country 599.864 30.653 347.409 25.266 207.020 15.639
Intercept Variance 105.132 4.096 105.132 4.096 105.132 4.096
ICC 0.149 0.232 0.337
Additional information
-2LL 246,386.630
N18,800
Taking care of business: the impact of culture and gender on entrepreneurs’ blended
123
Author's personal copy
deviation below the mean have economic goals which
are 1.1 times higher than their female counterparts.
Conversely, male entrepreneurs in societies with
levels post-materialism at one standard deviation over
the mean have economic goals that are 1.05 times
higher than their female counterparts. Figure 2b
presents the post-estimated values based on the
cross-level interaction models for social goals. Female
entrepreneurs in societies with post-materialism levels
at one standard deviation below the mean have
economic value goals which are 1.11 times higher
than their male counterparts. Conversely, female
entrepreneurs in societies with levels post-materialism
one standard deviation over the mean have economic
value goals that are 1.23 times higher than males.
In sum, our evidence supports H1a, H1b, H2a, H2b,
H2c, H3a, and H3b. Women entrepreneurs tend to
emphasize social value creation goals and deempha-
size economic value creation goals compared to men
entrepreneurs. Entrepreneurs in strong post-material-
ist cultures emphasize social and environmental value
creation goals over economic value creation goals.
Finally, there is a cross-level interaction between
gender and culture: post-materialism impacts the
Table 4 Simultaneous multi-level multivariate regression control model
Level and variable Economic Social Environmental
Estimate SE pvalue Estimate SE pvalue Estimate SE pvalue
Level 1
Intercept 77.329 11.505 0.000 25.231 6.431 0.000 -2.537 5.112 0.620
Age 0.010 0.018 0.601 -0.024 0.014 0.092 0.013 0.011 0.221
Household income 0.768 0.289 0.008 -0.572 0.219 0.009 -0.165 0.167 0.325
Education -0.887 0.181 0.000 0.587 0.137 0.000 0.280 0.105 0.008
Equal income -0.951 0.421 0.024 -0.217 0.319 0.496 1.172 0.244 0.000
Good career -0.084 0.459 0.855 -0.726 0.348 0.037 0.841 0.266 0.002
Status 0.720 0.470 0.126 0.090 0.357 0.800 -0.838 0.273 0.002
Media -0.777 0.443 0.080 0.137 0.336 0.682 0.694 0.257 0.007
Established business 3.154 0.435 0.000 -1.422 0.331 0.000 -1.751 0.253 0.000
Necessity 0.783 0.435 0.072 -0.833 0.330 0.012 0.082 0.253 0.747
Number of owners -0.007 0.003 0.010 0.001 0.002 0.736 0.006 0.001 0.000
Innovativeness -0.499 0.175 0.004 0.253 0.133 0.057 0.247 0.102 0.015
Male-dominated industry 0.274 1.030 0.790 -0.625 0.783 0.425 0.328 0.599 0.584
Extractive -2.026 0.735 0.006 -2.376 0.558 0.000 4.449 0.427 0.000
Transforming 1.787 0.541 0.001 -2.164 0.411 0.000 0.432 0.314 0.169
Business services 2.594 0.658 0.000 -1.529 0.498 0.002 -0.994 0.381 0.009
Level 2
GDP (per capita) 0.000 0.000 0.183 0.000 0.000 0.185 0.000 0.000 0.192
% GDP growth 0.234 0.335 0.485 -0.039 0.186 0.836 -0.196 0.153 0.200
% Unemployment 0.261 0.280 0.351 -0.258 0.155 0.095 -0.005 0.127 0.971
% Active labor force 0.152 0.143 0.288 -0.091 0.080 0.256 -0.063 0.065 0.329
% Tertiary education 0.075 0.081 0.355 -0.114 0.045 0.011 0.041 0.037 0.269
Gender stereotypes -10.551 3.872 0.006 4.843 2.135 0.023 5.707 1.760 0.001
Ecological footprint -0.505 1.431 0.724 0.509 0.795 0.522 -0.034 0.654 0.958
% TEA Necessity (2008) 0.630 0.688 0.360 -0.631 0.380 0.097 0.003 0.313 0.993
Variance components
Within-country 590.298 6.784 342.419 3.935 200.332 2.302
Within-individual 64.248 23.214 63.679 23.2405323 63.816 23.234
ICC 0.098 0.157 0.242
Additional information
-2LL 197,118.44
D-2LL 49,268.191
N15,141
D. M. Hechavarrı
´a et al.
123
Author's personal copy
relationship between gender and value creation goals
such that women’s goals change more notably as the
society shifts to post-materialism. In other words, a
post-materialistic culture amplifies the effects of
gender on value creation goals.
5.1 Robustness checks
To confirm the sensitivity of our results, we undertake
three follow-up analyses. First, we complete three
separate independent multi-level regression models on
Table 5 Simultaneous multi-level multivariate regression main effects model
Level and variable Economic Social Environmental
Estimate SE pvalue Estimate SE pvalue Estimate SE pvalue
Level 1
Intercept 72.736 11.997 0.000 24.708 2.017 0.000 -0.519 5.505 0.925
Age 0.008 0.018 0.659 -0.023 0.014 0.102 0.014 0.011 0.194
Household income 0.656 0.289 0.023 -0.477 0.219 0.030 -0.150 0.168 0.371
Education -0.881 0.181 0.000 0.580 0.137 0.000 0.281 0.105 0.007
Equal income -0.907 0.421 0.031 -0.267 0.319 0.402 1.180 0.244 0.000
Good career -0.137 0.459 0.765 -0.683 0.348 0.050 0.851 0.266 0.001
Status 0.744 0.470 0.114 0.044 0.357 0.902 -0.815 0.274 0.003
Media -0.775 0.443 0.080 0.143 0.335 0.670 0.687 0.257 0.008
Established business 3.086 0.435 0.000 -1.347 0.331 0.000 -1.761 0.253 0.000
Necessity 0.810 0.435 0.062 -0.847 0.330 0.010 0.068 0.253 0.787
Number of owners -0.006 0.003 0.011 0.001 0.002 0.756 0.006 0.001 0.000
Innovativeness -0.518 0.175 0.003 0.270 0.132 0.041 0.249 0.102 0.014
Male-dominated industry 0.300 1.029 0.770 -0.652 0.782 0.405 0.330 0.599 0.582
Extractive -2.209 0.736 0.003 -2.174 0.559 0.000 4.425 0.428 0.000
Transforming 1.570 0.542 0.004 -1.948 0.412 0.000 0.428 0.315 0.175
Business Services 2.377 0.659 0.000 -1.314 0.499 0.008 -0.997 0.382 0.009
Female -2.098 0.419 0.000 1.892 0.319 0.000 0.159 0.244 0.340
Level 2
GDP (per capita) 0.000 0.000 0.246 0.000 0.000 0.267 0.000 0.000 0.230
% GDP growth 0.295 0.339 0.384 -0.068 0.188 0.717 -0.228 0.154 0.140
% Unemployment 0.230 0.274 0.400 -0.232 0.152 0.127 -0.001 0.125 0.993
% Active labor force 0.180 0.139 0.197 -0.107 0.077 0.164 -0.075 0.063 0.237
% Tertiary education 0.059 0.079 0.458 -0.103 0.044 0.019 0.045 0.036 0.207
Gender stereotypes -8.190 3.768 0.030 3.248 2.069 0.116 4.950 1.702 0.004
Ecological footprint 0.340 1.609 0.833 -0.007 0.897 0.994 -0.362 0.736 0.623
% TEA necessity (2008) 0.684 0.652 0.294 -0.668 0.361 0.064 -0.014 0.296 0.961
% Post-materialism -0.299 0.277 0.050 0.193 0.153 0.027 0.105 0.126 0.044
Variance components
Within-country 587.168 6.748 339.792 3.905 200.747 2.307
Intercept variance 74.859 4.014 74.273 4.04 74.445 23.233
ICC 0.113 0.179 0.271
Additional information
-2LL 197,036.42
D-2LL 44,079.9
N15,141
Taking care of business: the impact of culture and gender on entrepreneurs’ blended
123
Author's personal copy
the full sample (see Table 7a). Again, we find gender
is significant for economic (b=-2.162; p\0.001)
and social (b=2.135; p\0.001) value goals, pro-
viding additional evidence to support the hypotheses.
However, post-materialism has no significant direct
effect on blended value creation goals. The model’s
lack of evidence for post-materialism’s direct effect is
likely due to the lack of correlated residuals in each
Table 6 Simultaneous multi-level multivariate regression cross-level interaction model
Level and variable Economic Social Environmental
Estimate SE pvalue Estimate SE pvalue Estimate SE pvalue
Level 1
Intercept 71.628 9.582 0.000 28.826 5.284 0.000 -0.430 4.245 0.919
Age 0.009 0.018 0.644 -0.023 0.014 0.096 0.014 0.011 0.193
Household income 0.660 0.289 0.022 -0.483 0.219 0.027 -0.148 0.168 0.377
Education -0.864 0.181 0.000 0.561 0.137 0.000 0.282 0.105 0.007
Equal income -0.873 0.421 0.038 -0.301 0.319 0.345 1.180 0.244 0.000
Good career -0.166 0.459 0.717 -0.657 0.348 0.059 0.853 0.267 0.001
Status 0.742 0.470 0.115 0.045 0.357 0.900 -0.814 0.274 0.003
Media -0.790 0.443 0.074 0.156 0.335 0.642 0.688 0.257 0.007
Established business 3.082 0.435 0.000 -1.342 0.330 0.000 -1.761 0.253 0.000
Necessity 0.797 0.435 0.067 -0.834 0.330 0.011 0.068 0.253 0.789
Number of owners -0.007 0.003 0.010 0.001 0.002 0.731 0.006 0.001 0.000
Innovativeness -0.524 0.174 0.003 0.276 0.133 0.037 0.249 0.101 0.014
Male-dominated industry 0.329 1.029 0.749 -0.680 0.782 0.384 0.330 0.599 0.582
Extractive -2.191 0.736 0.003 -2.196 0.559 0.000 4.429 0.428 0.000
Transforming 1.548 0.542 0.004 -1.928 0.412 0.000 0.430 0.315 0.173
Business Services 2.299 0.660 0.000 -1.239 0.499 0.013 -0.994 0.382 0.009
Female 0.512 0.876 0.559 -0.845 0.666 0.204 0.504 0.510 0.002
Level 2
GDP (per capita) 0.000 0.000 0.249 0.000 0.000 0.270 0.000 0.000 0.235
% GDP growth 0.303 0.337 0.367 -0.075 0.188 0.689 -0.229 0.153 0.133
% Unemployment 0.242 0.266 0.362 -0.240 0.148 0.104 -0.004 0.120 0.972
% Active labor force 0.181 0.137 0.185 -0.107 0.075 0.156 -0.076 0.061 0.216
% Tertiary education 0.059 0.080 0.461 -0.104 0.045 0.020 0.046 0.036 0.203
Gender stereotypes -8.184 3.529 0.020 3.264 1.952 0.095 4.927 1.580 0.002
Ecological footprint 0.297 1.564 0.849 0.021 0.869 0.981 -0.347 0.711 0.626
% TEA necessity (2008) 0.683 0.644 0.289 -0.662 0.358 0.064 -0.019 0.291 0.948
Post-materialism -0.229 0.269 0.394 0.118 0.149 0.429 0.111 0.122 0.731
Cross-level interaction
Female 9Post-Materialism -0.206 0.061 0.001 0.216 0.046 0.000 -0.012 0.035 0.731
Variance components
Residual variance 587.545 6.752 339.217 3.898 201.114 2.312
Intercept variance 83.397 4.494 82.757 2.859 83.037 2.852
ICC 0.124 0.196 0.292
Additional information
-2LL 197,034
D-2LL 2.42
N15,141
D. M. Hechavarrı
´a et al.
123
Author's personal copy
regression as the three variables (economic, social,
and environmental) are not independent of each other.
Therefore, estimating three separate regression equa-
tions fails to take this lack of independence into
account. Table 7b also reveals the significant interac-
tion effect for gender and post-materialism on eco-
nomic value (b=-0.207; p\0.001) and social
value (b=0.192; p\0.001) goals, again providing
further evidence to support the interaction claims of
our hypotheses. The individual coefficients and stan-
dard errors produced by our multi-level multivariate
regression system are similar to those produced by
separately estimating each independent multi-level
regression. However, the key difference is that simul-
taneous multi-level multivariate regression, being a
joint estimator, also estimates the between-equation
covariance, which separate multi-level regressions do
not. This is likely why we see subtle differences in the
significance values associated with post-materialism
at the societal level in the main effects model.
Second, we complete three sub-sample ordinary
least squares (OLS) regression on the top 75 % of
countries on the level of GDP per capita (PPP) (cut-off
is 32,000, yielding 14 countries).
6
Since we have ten
country-level variables, we need at least 36 second
level units for a multi-level regression model (Bell
et al. 2010); consequently, a multi-level approach for
our sub-sample analysis is inappropriate. Also, the
sub-sample is so similar that the ICC scores are under
critical threshold of 0.08 (Aguinus and Culpepper
2015). The OLS sub-sample findings also provide
evidence to support Hypotheses 1a and 1b: women
entrepreneurs prioritize economic value less
(b=-3.528; p\0.001) than social value
(b=3.514; p\0.001), compared to men entrepre-
neurs (see Table 8a). The OLS sub-sample findings
also support Hypotheses 2b, in societies with high
levels of GDP, the level of post-materialism signifi-
cantly impacts social value goals (b=0.423;
p=0.001). Finally, Table 8b illustrates that, as
post-materialism levels increase, women entrepre-
neurs’ emphasis on economic value creation goals
decreases (b=-0.378; p=0.005), but social value
goals increase (b=0.345; p=0.001).
We complete another set of sub-sample OLS
regressions on the lowest 25 % of countries on the
level of GDP PPP (cut-off =less than 8000; yields 8
countries).
7
These OLS sub-sample findings also
support Hypotheses 1a and 1b: women entrepreneurs
a
b
Post-materialism
Post-materialism
Fig. 2 Predicted marginal
means of men and women
in different post-materialist
cultural contexts for
economic value goals
(a) and social value goals
(b). Note: Covariates are
estimated at the means to
generate estimated
marginal effects
6
These countries are: United States, Netherlands, Belgium,
France, Spain, Italy, Switzerland, United Kingdom, Denmark,
Norway, Germany, Iceland, Finland, and Hong Kong.
7
These countries are Argentina, Morocco, Algeria, Uganda,
Guatemala, Jordan, Yemen, and West Bank and Gaza Strip.
Taking care of business: the impact of culture and gender on entrepreneurs’ blended
123
Author's personal copy
Table 7 Separate multi-level regression models: (a) main effect and (b) interaction effects
Level and variable Economic
Random intercept and fixed
slope
Social
Random intercept and fixed
slope
Environmental
Random intercept and fixed
slope
Estimate SE pEstimate SE pEstimate SE p
(a)
Level 1
Intercept 72.149 15.011 0.000 28.932 10.288 0.005 0.600 7.428 0.936
Age 0.007 0.019 0.019 -0.011 0.014 0.451 0.003 0.011 0.776
Household income 0.624 0.291 0.291 -0.260 0.220 0.236 -0.298 0.168 0.076
Education -0.903 0.184 0.184 0.793 0.139 0.000 0.126 0.106 0.236
Equal income -0.977 0.426 0.426 0.114 0.322 0.724 0.810 0.246 0.001
Good career -0.194 0.464 0.464 -0.459 0.351 0.191 0.577 0.268 0.031
Status 0.769 0.474 0.474 -0.057 0.358 0.873 -0.783 0.274 0.004
Media -0.851 0.448 0.448 0.371 0.339 0.275 0.501 0.259 0.053
Established business 3.141 0.437 0.437 -1.490 0.330 0.000 -1.567 0.253 0.000
Necessity 0.747 0.438 0.438 -0.541 0.331 0.102 -0.222 0.253 0.380
Number of owners -0.006 0.003 0.003 0.000 0.002 0.820 0.006 0.001 0.000
Innovativeness -0.523 0.175 0.175 0.267 0.132 0.043 0.211 0.101 0.038
Male-dominated ind. 0.161 1.035 1.035 0.275 0.777 0.724 -0.358 0.598 0.550
Extractive -2.185 0.741 0.741 -2.397 0.557 0.000 4.497 0.428 0.000
Transforming 1.535 0.547 0.547 -1.932 0.412 0.000 0.304 0.316 0.335
Business services 2.270 0.669 0.669 -0.736 0.504 0.144 -1.473 0.386 0.000
Female -2.162 0.421 0.000 2.135 0.318 0.000 -0.043 0.243 0.859
Level 2
GDP (per capita) 0.000 0.000 0.000 0.000 0.000 0.939 0.000 0.000 0.228
% GDP growth 0.291 0.341 0.341 -0.045 0.233 0.847 -0.254 0.168 0.132
% Unemployment 0.253 0.287 0.287 -0.256 0.197 0.194 -0.019 0.142 0.892
% Active labor force 0.186 0.170 0.170 -0.101 0.117 0.387 -0.083 0.084 0.322
% Tertiary education 0.056 0.084 0.084 -0.079 0.058 0.174 0.035 0.042 0.403
Gender stereotypes -7.952 4.737 4.737 1.772 3.243 0.585 5.856 2.339 0.012
Ecological footprint 0.721 0.695 0.695 -0.763 0.476 0.109 -0.007 0.343 0.983
% TEA necessity (2008) 0.079 1.694 1.694 -0.202 1.164 0.863 -0.180 0.842 0.831
Post-materialism -0.312 0.295 0.295 0.312 0.202 0.122 0.052 0.146 0.721
Variance components
Within-country 72.214 0.898 33.514 0.424 17.313 0.225
Within-individual 585.125 0.898 328.597 0.011 195.335 0.006
Additional information
ICC 0.110 0.093 0.081
-2LL -69,795.177 -64,339.594 -61,487.031
Pseudo R
2
0.300 0.290 0.310
N15,140 15,140 15,140
(b)
Level 1
Intercept 71.095 15.038 0.000 29.892 10.300 0.004 0.633 7.432 0.932
Age 0.007 0.019 0.698 -0.011 0.014 0.433 0.003 0.011 0.777
Household income 0.628 0.291 0.031 -0.264 0.220 0.230 -0.299 0.168 0.076
D. M. Hechavarrı
´a et al.
123
Author's personal copy
significantly rate economic value lower (b=-1.638;
p=0.019) than social value (b=1.535; p=0.008),
when compared to men entrepreneurs (Table 9a).
However, the OLS sub-sample findings also indicate
that in poorer countries, as post-materialism increases,
social value goals (b=-2.121; p=0.001) decrease,
and environmental value goals (b=-4.504;
p\0.0001) decrease, yet economic goals increase
(b=6.568; p\0.0001). This is contrary to the
effect of post-materialism in the overall sample.
There is no significant interaction among the value
goals in this segment of the sub-sample. All results
from these additional analyses are available upon
request.
Table 7 continued
Level and variable Economic
Random intercept and fixed
slope
Social
Random intercept and fixed
slope
Environmental
Random intercept and fixed
slope
Estimate SE pEstimate SE pEstimate SE p
Education -0.884 0.184 0.000 0.775 0.139 0.000 0.125 0.106 0.239
Equal income -0.940 0.426 0.027 0.080 0.322 0.805 0.808 0.246 0.001
Good career -0.223 0.464 0.632 -0.430 0.351 0.221 0.578 0.268 0.031
Status 0.766 0.474 0.106 -0.054 0.358 0.881 -0.783 0.274 0.004
Media -0.864 0.448 0.054 0.383 0.339 0.258 0.501 0.259 0.053
Established business 3.135 0.437 0.000 -1.483 0.330 0.000 -1.567 0.253 0.000
Necessity 0.735 0.438 0.093 -0.533 0.331 0.107 -0.222 0.253 0.381
Number of owners -0.007 0.003 0.011 0.000 0.002 0.802 0.006 0.001 0.000
Innovativeness -0.529 0.175 0.003 0.272 0.132 0.039 0.211 0.101 0.037
Male-dominated industry 0.194 1.034 0.851 0.247 0.777 0.750 -0.359 0.598 0.548
Extractive -2.168 0.741 0.003 -2.411 0.557 0.000 4.497 0.428 0.000
Transforming 1.513 0.546 0.006 -1.908 0.411 0.000 0.305 0.316 0.334
Business services 2.193 0.669 0.001 -0.664 0.504 0.187 -1.471 0.387 0.000
Female 0.463 0.880 0.599 -0.278 0.663 0.675 -0.129 0.508 0.799
Level 2
GDP(per capita) 0.000 0.000 0.469 0.000 0.000 0.978 0.000 0.000 0.229
% GDP growth 0.299 0.342 0.382 -0.052 0.234 0.823 -0.254 0.168 0.131
% Unemployment 0.264 0.288 0.359 -0.266 0.197 0.178 -0.020 0.142 0.891
% Active labor force 0.187 0.171 0.273 -0.102 0.117 0.382 -0.083 0.084 0.322
% Tertiary education 0.056 0.085 0.508 -0.079 0.058 0.174 0.035 0.042 0.403
Gender stereotypes -7.961 4.744 0.093 1.781 3.246 0.583 5.856 2.340 0.012
Ecological footprint 0.717 0.696 0.303 -0.758 0.476 0.111 -0.007 0.343 0.983
% TEA necessity (2008) 0.048 1.697 0.977 -0.172 1.166 0.882 -0.179 0.842 0.832
Post-materialism -0.240 0.296 0.418 0.245 0.203 0.227 0.050 0.146 0.734
Cross-level interaction
Female 9post-materialism -0.207 0.061 0.001 0.192 0.046 0.001 0.007 0.035 0.847
Variance components
Within-country 72.464 6.782 33.582 3.871 2.060 2.253
Within-individual 584.672 23.873 328.215 10.801 246.394 5.791
Additional information
ICC 0.011 0.093 0.008
-2LL -69,789.408 -64,339.594 -23,014.901
Pseudo R
2
0.360 0.320 0.340
N15,140 15,140 15,140
Taking care of business: the impact of culture and gender on entrepreneurs’ blended
123
Author's personal copy
Table 8 Ordinary least squares sub-sample analysis: 75th percentile of post-materialism: (a) main effects and (b) interaction effects
Level and variable OLS economic OLS social OLS environmental
Estimate SE pEstimate SE pEstimate SE p
(a)
Intercept 71.020 17.097 0.000 5.464 13.091 0.676 19.262 10.286 0.061
Age 0.011 0.035 0.766 -0.019 0.027 0.489 0.012 0.021 0.577
Household income 2.155 0.537 0.000 -0.714 0.416 0.086 -1.242 0.323 0.000
Education -0.734 0.325 0.024 1.172 0.251 0.000 -0.390 0.195 0.046
Equal income -1.146 0.730 0.116 0.396 0.565 0.482 0.705 0.439 0.108
Good career -0.331 0.741 0.655 -0.674 0.573 0.240 0.824 0.446 0.064
Status 0.719 0.773 0.352 -0.088 0.597 0.882 -0.839 0.465 0.071
Media -1.430 0.745 0.055 0.630 0.577 0.275 0.918 0.448 0.041
Established business 4.069 0.772 0.000 -2.149 0.597 0.000 -1.840 0.464 0.000
Necessity 0.132 0.874 0.880 -0.300 0.676 0.657 0.284 0.526 0.590
Number of owners -0.011 0.005 0.038 0.000 0.004 0.985 0.011 0.003 0.001
Innovativeness -0.886 0.334 0.008 0.316 0.258 0.219 0.451 0.201 0.025
Male-dominated Ind. -3.340 2.389 0.162 -0.122 1.832 0.947 3.491 1.437 0.015
Extractive -5.402 1.439 0.000 -3.919 1.104 0.000 8.998 0.866 0.000
Transforming 3.255 0.999 0.001 -3.534 0.771 0.000 0.044 0.601 0.941
Business services 5.178 1.058 0.000 -3.306 0.815 0.000 -2.095 0.636 0.001
Female -3.528 0.754 0.000 3.515 0.583 0.000 -0.200 0.454 0.659
GDP (per capita) 0.000 0.000 0.535 0.000 0.000 0.200 0.000 0.000 0.774
% GDP growth 1.368 0.562 0.015 -1.219 0.430 0.005 -0.184 0.338 0.586
% Unemployment 0.211 0.314 0.503 0.021 0.241 0.931 -0.060 0.189 0.753
% Active labor force 0.105 0.114 0.356 -0.059 0.087 0.501 -0.064 0.069 0.352
% Tertiary education -0.172 0.066 0.009 0.114 0.050 0.023 0.044 0.039 0.265
Gender stereotypes 0.707 2.208 0.749 0.008 1.687 0.996 -0.415 1.328 0.755
Ecological footprint 0.998 0.594 0.093 -2.406 0.457 0.000 0.575 0.357 0.108
% TEA necessity (2008) 8.550 2.795 0.002 -9.060 2.145 0.000 -1.490 1.682 0.376
Post-materialism -0.259 0.171 0.131 0.423 0.132 0.001 0.067 0.103 0.514
R
2
0.047 0.062 0.045
F11.400 14.190 10.370
pvalue 0.000 0.000 0.000
N5513 5513 5513
(b)
Intercept 63.653 17.272 0.000 17.275 13.184 0.019 17.628 10.399 0.090
Age 0.012 0.035 0.741 -0.020 0.027 0.452 0.012 0.021 0.570
Household income 2.186 0.537 0.000 -0.740 0.415 0.075 -1.246 0.323 0.000
Education -0.714 0.325 0.028 1.134 0.250 0.000 -0.385 0.195 0.049
Equal income -0.893 0.738 0.226 -0.130 0.570 0.819 0.803 0.444 0.071
Good career -0.474 0.742 0.523 -0.440 0.572 0.442 0.795 0.446 0.075
Status 0.683 0.772 0.376 -0.037 0.595 0.951 -0.844 0.465 0.069
Media -1.411 0.745 0.058 0.564 0.575 0.326 0.939 0.449 0.036
Established Business 4.085 0.771 0.000 -2.185 0.595 0.000 -1.827 0.464 0.000
Necessity 0.097 0.874 0.912 -0.264 0.674 0.695 0.281 0.526 0.593
Number of owners -0.011 0.005 0.038 0.000 0.004 0.912 0.010 0.003 0.001
D. M. Hechavarrı
´a et al.
123
Author's personal copy
6 Discussion
Prior research generally examines the career choice
of individuals into entrepreneurship (Parker 2009), or
social entrepreneurship (Estrin et al. 2013), but does
not simultaneously consider the blended value goals
of various kinds of entrepreneurs—the key contribu-
tion of this study. We address a literature gap around
how entrepreneurs combine and prioritize the differ-
ent types of values they aim to create with their
ventures. Multi-level multivariate regression method-
ology enables us to examine multiple levels of
variation among our dependent variables with con-
temporaneously correlated errors. We analyze the
value creation goals of women and men entrepre-
neurs across individuals and countries. Our study
extends and refines recent studies analyzing national
values’ impact on individual entrepreneurs (e.g.,
Reynolds 2010; Shinnar et al. 2012) by examining
how these values interact with gender. Our findings
suggest that gender significantly influences entrepre-
neurs’ value creation preferences: women are less
likely to emphasize economic value, and more likely
to emphasize social value. We believe this finding
can be attributed to tangible differences in how men
and women frame ethical decisions (Jaffee and Hyde
2000). However, our findings do not provide suffi-
cient evidence to suggest that gender impacts
entrepreneurs’ environmental value creation targets.
We also find evidence of a direct effect of post-
materialism on entrepreneurs’ social, economic, and
environmental value creation goals, highlighting the
important role of normative values on organizations’
blended value creation. Finally, we highlight the
importance of cultural embeddedness as evidenced
by the cross-level interactions between gender and
level of post-materialism in terms of economic and
social value creation goals. Specifically, as countries
move toward higher levels of post-materialism, the
relationship between gender and social as well as
economic value creation goals becomes stronger. The
gender gap in social and economic value creation
goals is wider in post-materialistic societies than in
materialistic ones.
Table 8 continued
Level and variable OLS economic OLS social OLS environmental
Estimate SE pEstimate SE pEstimate SE p
Innovativeness -0.928 0.335 0.006 0.400 0.257 0.120 0.434 0.202 0.031
Male-dominated Ind. -3.219 2.388 0.178 -0.123 1.827 0.946 3.431 1.438 0.017
Extractive -5.377 1.441 0.000 -3.730 1.103 0.001 8.892 0.868 0.000
Transforming 3.225 1.002 0.001 -3.313 0.771 0.000 -0.040 0.603 0.947
Business services 4.886 1.068 0.000 -2.638 0.821 0.001 -2.239 0.643 0.001
Female 3.214 2.550 0.208 -2.487 1.955 0.203 -1.104 1.535 0.472
GDP (per capita) 0.000 0.000 0.629 0.000 0.000 0.334 0.000 0.000 0.836
% GDP growth 1.408 0.562 0.012 -1.325 0.429 0.002 -0.159 0.338 0.638
% Unemployment 0.303 0.317 0.340 -0.162 0.242 0.502 -0.020 0.191 0.915
% Active labor Force 0.095 0.114 0.405 -0.022 0.087 0.800 -0.075 0.069 0.275
% Tertiary education -0.250 0.080 0.002 0.300 0.061 0.000 0.000 0.048 0.994
Gender stereotypes 2.483 2.448 0.311 -4.279 1.868 0.022 0.615 1.474 0.677
Ecological footprint 8.196 2.802 0.003 -8.278 2.143 0.000 -1.706 1.687 0.312
% TEA necessity (2008) 1.017 0.594 0.087 -2.425 0.456 0.000 0.573 0.357 0.109
Post-materialism -0.157 0.180 0.383 0.393 0.138 0.004 0.029 0.108 0.788
Female 9post-materialism -0.378 0.135 0.005 0.345 0.104 0.001 0.048 0.081 0.556
R
2
0.049 0.064 0.045
F10.950 14.090 9.980
pvalue 0.000 0.000 0.000
N5513 5513 5513
Taking care of business: the impact of culture and gender on entrepreneurs’ blended
123
Author's personal copy
Table 9 Ordinary least squares sub-sample analysis: 25th percentile of post-materialism: (a) main effects and (b) interaction effects
Level and variable OLS economic OLS social OLS environmental
Estimate SE pEstimate SE pEstimate SE p
(a)
Intercept -303.559 31.112 0.000 190.201 25.821 0.000 218.356 14.577 0.000
Age -0.003 0.029 0.915 0.016 0.024 0.525 -0.014 0.014 0.313
Household income 0.649 0.449 0.148 -0.719 0.373 0.054 0.074 0.210 0.723
Education -0.873 0.325 0.007 0.606 0.270 0.025 0.286 0.152 0.060
Equal income 1.777 0.721 0.014 -1.360 0.600 0.023 -0.392 0.338 0.246
Good career 1.585 0.880 0.072 -1.098 0.733 0.134 -0.605 0.412 0.143
Status 1.642 0.944 0.082 -0.865 0.786 0.271 -0.841 0.442 0.057
Media -1.964 0.880 0.026 1.401 0.731 0.055 0.654 0.412 0.113
Established business 0.331 0.750 0.659 0.272 0.623 0.662 -0.559 0.351 0.112
Necessity 0.371 0.692 0.591 -0.285 0.575 0.620 -0.061 0.324 0.850
Number of owners -0.002 0.004 0.704 -0.003 0.003 0.360 0.005 0.002 0.016
Innovativeness -0.031 0.281 0.913 -0.064 0.233 0.784 0.062 0.132 0.639
Male-dominated Ind. 0.093 1.468 0.950 1.057 1.222 0.387 -1.021 0.688 0.138
Extractive 0.788 1.007 0.434 -2.238 0.836 0.007 1.377 0.472 0.004
Transforming 1.373 0.952 0.149 -1.096 0.791 0.166 -0.307 0.446 0.491
Business services -1.101 1.465 0.452 0.101 1.217 0.934 1.091 0.686 0.112
Female -1.638 0.698 0.019 1.535 0.580 0.008 0.165 0.327 0.615
GDP (per capita) 0.014 0.001 0.000 -0.006 0.001 0.000 -0.008 0.001 0.000
% GDP growth -5.332 0.639 0.000 2.493 0.530 0.000 2.911 0.300 0.000
% Unemployment 9.012 0.685 0.000 -4.284 0.569 0.000 -4.889 0.321 0.000
% Active labor Force 3.262 0.293 0.000 -1.208 0.243 0.000 -2.083 0.137 0.000
% Tertiary education 0.836 0.063 0.000 -0.479 0.053 0.000 -0.320 0.030 0.000
Gender stereotypes -166.471 13.549 0.000 61.231 11.240 0.000 103.879 6.348 0.000
Ecological footprint -61.347 4.993 0.000 22.564 4.142 0.000 38.281 2.339 0.000
% TEA necessity (2008) 13.595 1.098 0.000 -7.182 0.911 0.000 -6.511 0.514 0.000
Post-materialism 6.568 0.539 0.000 -2.121 0.447 0.000 -4.504 0.253 0.000
R
2
0.218 0.115 0.243
F47.800 30.770 54.930
pvalue 0.000 0.000 0.000
N4132 4132 4132
(b)
Intercept -304.688 31.208 0.000 191.108 25.901 0.000 -80.658 14.622 0.000
Age -0.003 0.029 0.917 0.015 0.024 0.527 -0.014 0.014 0.313
Household income 0.645 0.449 0.151 -0.715 0.373 0.055 0.075 0.210 0.720
Education -0.872 0.325 0.007 0.606 0.270 0.025 0.285 0.152 0.061
Equal income 1.776 0.721 0.014 -1.359 0.600 0.024 -0.392 0.338 0.247
Good career 1.586 0.880 0.072 -1.099 0.733 0.134 -0.605 0.412 0.143
Status 1.632 0.945 0.084 -0.857 0.786 0.276 -0.839 0.443 0.058
Media -1.971 0.880 0.025 1.407 0.731 0.054 0.655 0.412 0.112
Established business 0.320 0.750 0.670 0.281 0.623 0.652 -0.556 0.352 0.114
Necessity 0.377 0.692 0.586 -0.290 0.575 0.615 -0.062 0.324 0.847
Number of owners -0.001 0.004 0.713 -0.003 0.003 0.355 0.005 0.002 0.016
D. M. Hechavarrı
´a et al.
123
Author's personal copy
On average, globally, women favor social goals
over economic goals. For example, in the USA,
Norway, Netherlands, and Japan, women’s social goal
average is 33 %, compared to 26 % for men. In
addition, our results show that countries with citizens
that express higher levels of post-materialist values are
also, on average, associated with higher social and
environmental value goals. For example, in the
Netherlands, Uruguay, and Norway, which are rela-
tively high on post-materialism, entrepreneurs’ social
goal average is 24.3 %, compared to 6.9 % in Russia,
Serbia, Hungary (see Table 10). Since women are
more strongly affected by post-materialism than men,
women entrepreneurs’ social goals in the USA and
Norway actually average 33 %, which is 10 % higher
than social goals of men in these same countries. At
the same time, women entrepreneurs’ social goals in
Greece and Switzerland average 25.5 % which is only
5 % higher than social goals of men in these countries.
With these empirical results, we offer four contri-
butions to previous research on gender, culture, and
entrepreneurship. First, previous studies on female
entrepreneurship often emphasize how women’s’
venture goals and aspirations differ from men’s,
although the reasons for these gender differences were
poorly understood (de Bruin et al. 2006; Hughes et al.
2012). Our study introduces the ethics of care and
ethics of justice perspectives to explain these differ-
ences. Ethics of care emphasizes socialization, thus
complementing previous studies on women
Table 9 continued
Level and variable OLS economic OLS social OLS environmental
Estimate SE pEstimate SE pEstimate SE p
Innovativeness -0.032 0.281 0.908 -0.062 0.233 0.789 0.062 0.132 0.637
Male-dominated Ind. 0.114 1.469 0.938 1.040 1.223 0.395 -1.026 0.688 0.136
Extractive 0.776 1.007 0.441 -2.228 0.836 0.008 1.380 0.472 0.003
Transforming 1.364 0.953 0.152 -1.088 0.791 0.169 -0.306 0.446 0.494
Business services -1.117 1.465 0.446 0.113 1.218 0.926 1.094 0.687 0.111
Female -1.128 1.293 0.383 1.126 1.073 0.294 0.059 0.606 0.923
GDP (per capita) 0.014 0.001 0.000 -0.006 0.001 0.000 -0.008 0.001 0.000
% GDP growth -5.349 0.640 0.000 2.507 0.531 0.000 2.915 0.300 0.000
% Unemployment 9.034 0.686 0.000 -4.302 0.570 0.000 -4.893 0.322 0.000
% Active labor Force 3.272 0.294 0.000 -1.216 0.244 0.000 -2.085 0.138 0.000
% Tertiary education 0.838 0.063 0.000 -0.481 0.053 0.000 -0.321 0.030 0.000
Gender stereotypes -166.900 13.581 0.000 61.576 11.267 0.000 103.968 6.364 0.000
Ecological footprint -61.505 5.005 0.000 22.692 4.152 0.000 38.314 2.345 0.000
% TEA necessity (2008) 13.627 1.100 0.000 -7.207 0.912 0.000 -6.518 0.515 0.000
Post-materialism 6.613 0.548 0.000 -2.158 0.455 0.000 -4.513 0.257 0.000
Female 9post-materialism -0.071 0.151 0.640 0.057 0.125 0.650 0.015 0.071 0.835
R
2
0.214 0.153 0.243
F45.89 29.54 52.73
pvalue 0.000 0.000 0.000
N4132 4132 4132
Table 10 Average reported blended value goals for post-materialism levels
Economic Social Environmental
Post-materialism (-)1SD (e.g., Russia, Serbia, Hungary) 85.2 6.9 7.8
Post-materialism average (e.g., Israel, Croatia, Slovenia) 58.7 23.25 18.13
Post-materialism (?)1SD (e.g., Netherlands, Uruguay, Norway) 43.25 24.35 14.06
Taking care of business: the impact of culture and gender on entrepreneurs’ blended
123
Author's personal copy
entrepreneurship showing how social processes lead to
differences in the kinds of ventures pursued between
genders (Gupta et al. 2009). Despite the recent interest
in ethics of care in management (Lawrence and Maitlis
2012), the literature generally focuses on how orga-
nizations are managed rather than why organizations
are initially created. Therefore, we provide evidence
for the role of gender in the earliest phases of firm
creation in terms of blended value goals.
Second, our study expands previous research on
post-materialism and entrepreneurship (Stephan et al.
2014; Uhlaner and Thurik 2004, 2007; Uhlaner et al.
2002) by investigating how post-materialism impacts
entrepreneurs’ blended value creation goals. As mod-
ernization moves countries toward a different set of
life goals (Inglehart 1977), entrepreneurs’ business
goals shift from financial profits to social and ecolog-
ical value creation outcomes. Indeed, in societies
where basic material and physical human needs are
largely met, entrepreneurs place less emphasis on the
economic value creation. However, this presents a
dilemma for policy in that socially focused ventures
are required when governments cannot fully manage
complex social and ecological issues and challenges
(Dacin et al. 2010) and therefore fulfill institutional
voids (Stephan et al. 2014), especially in less-devel-
oped, materially focused societies. Paradoxically,
however, entrepreneurs in such societies predomi-
nantly focus on pursuing economic value creation, not
social or environmental value creation. In light of our
findings, governments of weak post-materialistic
countries should understand the challenges that their
cultures pose for entrepreneurs’ social value creation
and perhaps encourage social venturing from foreign
sources to solve the myriad of social and ecological ills
and needs.
Third, the impact of gender on the venturing goals
depends on the cultural environment where
entrepreneurship takes place. Even if previous
research suggests that, compared to men, women’s
entrepreneurial activities may be more strongly
affected by cultural forces (de Bruin et al. 2007; Elam
2008; Bullough et al. 2014), theorizing and empirical
evidence on this topic is only beginning to emerge.
Here, our study makes an important contribution by
showing that socialization to one’s gender, as well as
to one’s culture, is reflected in the blended value
creation goals that entrepreneurs set for their new
ventures. Women are particularly quick to adapt
in situations where their gendered goals of caring
and nurturing are considered a weakness in a particular
domain, such as entrepreneurship in highly material-
istic societies. Instead of building organizations that
reflect the goals of compassion and communal values,
women entrepreneurs in weak post-materialistic soci-
eties tend to emphasize venturing goals that are closer
to those of men—that is, more economic. In strong
post-materialistic cultures that value humanism, qual-
ity of life, peace, human rights, and the environment,
women’s socialization to ethics of care is strongly
reflected in venturing goals. In such cultures, women
‘dare to care’’ and report significantly higher social
venturing goals than men in similar cultures, or
women (and men) in more materialistic cultures.
Here, our results align with prior research highlighting
contextual influences in ethical decision making, and
women’s heightened sensitivity (Christie et al. 2003;
Dalton and Ortegren 2011; Radtke 2000). We also
respond to calls to deepen our understanding of
gendered patterns using blended value creation frame-
work (Jennings and Brush 2013; Braun 2010) and
context (Welter 2011).
Finally, when investigating entrepreneurial activity
using the GEM research program, many studies use
GEM’s pre-defined and calculated measures, particu-
larly total entrepreneurial activity (TEA) which cap-
tures nascent entrepreneurs and young business
owners. While this approach facilitates replication
and cross-cultural comparison, there is seldom a
theoretical justification for selecting a particular
dependent variable (Bergmann et al. 2014). Our
research incorporates the APS micro-data to identify
potential variations in individual entrepreneurs’
blended value creation goals. The burgeoning focus
on blended value reflects the increased attention to
alternative conceptualizations of value creation and
the significant advantages firms reap by being green
and socially conscious (Bugg-Levine and Emerson
2011; Zahra et al. 2014). We contribute to this growing
field by leveraging the GEM database.
Our findings should be considered in light of
several limitations. First, there are no direct measures
for ethic of care in GEM data; therefore, we use
respondent’s gender as proxy. A meta-analysis of
studies using the ‘‘ethic of care inventory’’ reports that
women generally score higher than men in care ethics
(Jaffee and Hyde 2000). In our case, although we use
gender as a proxy for ethics of care and justice, our
D. M. Hechavarrı
´a et al.
123
Author's personal copy