Content uploaded by Hannah Stoppelmann
Author content
All content in this area was uploaded by Hannah Stoppelmann on Aug 13, 2019
Content may be subject to copyright.
Examining the Glass Ceiling for Female
Entrepreneurs: An Empirical Analysis
Hannah Stoppelmann
University of California, Los Angeles
Social Science Interdisciplinary Program
June 2019
Abstract
Extensive research on the glass ceiling has shown that the large gap between the female
representation at lower and upper levels of management is largely driven by differences in family
responsibilities, career interruptions and the requirement of working long hours. Additionally,
recent data suggests that female entrepreneurial activity lags behind that of males in terms of
business creation and revenue generation. Though one might expect female entrepreneurs to
bypass the glass ceiling constraints when becoming a business owner, research still documents
gender gaps in this market. This paper examines whether the traditional glass ceiling inhibitors
apply to female entrepreneurs as well. We find that male and female reach early levels of success
at similar rates and have similar aspirations for financial growth. However, we find that female
entrepreneurs are more likely to cite flexibility as an important reason for entering
entrepreneurship and spend more time on housekeeping and childcare.
1
1. Introduction
As members of the workforce, women have experienced significant changes since the
mid twentieth century. From 1948 to 2016, women’s share of the United States’ civilian labor
force grew by 70% (United States Department of Labor, 2016) and as of mid-2018, 44.7% of
employees of S&P 500 companies were women (Catalyst, 2018). Though we have seen progress
in female workforce participation, women are still vastly underrepresented in executive
leadership positions (Lyness & Grotto, 2018). As of 2018, only 26.5% of senior-level executive
positions and 4.8% of CEO positions in S&P 500 companies are held by women (Catalyst,
2018). This phenomenon is known as the glass ceiling—the idea that there is an invisible and
unreachable barrier that keeps minorities and women from rising to the top levels in their
profession regardless of their qualifications (Glass Ceiling Commission, 1995).
Just as the traditional labor force has seen an increase in female participation, the
entrepreneurship market has seen a similar trend. Though one may think of an entrepreneur as a
young, tech-focused, and venture-backed CEO, entrepreneurs come in many forms.
Entrepreneurship in general is actually defined as the pursuit of an opportunity beyond current
available and controlled resources (Eisenmann, 2013) This paper focuses on looking at the
general American entrepreneur, someone who aspires to be a business owner in a variety of
industries.
Overall, the share of female entrepreneurs has increased since the early 2000s. In fact, the
number of female-owned businesses – firms where the majority of equity, interest, or stock of
the business is owned by a woman – have grown at a faster rate than male-owned business. From
2002 to 2012, female-owned businesses in the U.S. increased by 35.6% compared to male-
owned firms which increased by 12.1% (National Women’s Business Council, 2012; U.S.
2
Department of Commerce, 2016). Despite this growth, the total share of new businesses started
by women is still lagging behind those started by men. As of 2015, the share of female-owned
businesses in the U.S. reached 36% compared with 55% for male-owned business and 9% for
equally-owned business (firms where the equity, interest, or stock of the business is shared 50-50
among men and women owners) (McManus 2017).
Becoming an entrepreneur might be expected to provide a way for women to bypass the
glass ceiling, allowing women to avoid getting stuck lower on the corporate ladder by starting at
the top level as a business owner. However, a gender gap remains when it comes to the
performance of these businesses. A trend parallel to the glass ceiling effect is evident in the
entrepreneurial market. Despite an increase in the number of new businesses owned by women,
we see a much larger gap between revenues generated by female-owned businesses versus
revenues generated by male-owned businesses. Looking at data from the U.S. Census Survey of
Business Owners, as of 2012, female-owned businesses made up 12% of all U.S. business
revenues compared to male-owned businesses which represented 79% (McManus, 2017).
In this paper, we investigate to what extent the factors that contribute to the glass ceiling
for women in the traditional labor market, also negatively affect women in the entrepreneurial
market. The next section reviews evidence of gender differences in these two markets to
understand why women may not reach the same levels of career performance as men. The
literature has found evidence that women’s roles in the home have been perpetuated by gender
social stereotypes (Eagly & Wood, 2016) and the idea that family well-being often leads to men
to focus more on financial support than women. (Becker, 1965). Despite an increase in male
participation in domestic responsibilities, women still take on a majority of these duties (Pew
Research Center, 2013). This disparity causes women to often prefer careers with more
3
flexibility, which can prevent many from rising to the top levels, as those positions often require
long hours and can have inflexible schedules (Goldin, 2014). The literature examining gender
differences among entrepreneurs is less robust but has found some distinct variation. Female
entrepreneurs are more likely to create necessity-based businesses than opportunity-based,
meaning that they decide to become an entrepreneur due to an employment necessity verses a
desire to pursue a new opportunity (Zwan, Thurik, Verheul, & Hessels, 2016). Women rank
balancing a family as an important factor when deciding to become an entrepreneur more often
than men, leading to a higher share of necessity-based businesses for women (Buttner & Moore,
1997; DeMartino & Barbato, 2003). Necessity-based businesses commonly underperform
opportunity-based businesses, contributing to a gender performance gap (Calderon, Iacovone, &
Juarez, 2017). Data on financial funding for entrepreneurs, often an important aspect of business
growth, indicates that female owners receive less traditional credit and equity financing than
male owners (Brush, Carter, Gatewood, Greene, & Hart, 2004; Carter & Rosa, 1998; Zarya,
2018). Recent research has found that investors treat female entrepreneurs differently than males
during their pitches, ultimately negatively influencing the amount of investment the women
receive (Alsos, Isaksen, & Ljunggren, 2006; Kanze, Huang, Conley, & Higgins, 2018).
This paper advances the literature by empirically studying how various individual and
firm characteristics impact female entrepreneurs’ success and financial aspirations compared to
male entrepreneurs. The Institute for Social Research at University of Michigan conducted a
five-year longitudinal survey from 2005 to 2010, called the Panel of Study of Entrepreneurial
Dynamics II, where they surveyed nascent (aspiring) entrepreneurs to understand the firm
creation process. The initial survey includes 1,214 nascent entrepreneurs. Overall, the individuals
in the sample are majority Caucasian and U.S born, have a large age distribution, are educated,
4
have a significant amount of work experience and have previous entrepreneurial experience. The
outcomes of interest that we analyze are one-year survival of an operational business, which
serves as a proxy for entrepreneurial success and one-year and five-year annual revenue
expectations, which serve as proxies for financial performance. Our empirical strategy focuses
on analyzing gender differences referenced in the literature such as business type (opportunity-
based vs. necessity-based), domestic responsibilities, motivations and funding. We estimate a
logistic regression to see how gender impacts the likelihood of one-year firm survival and a
quantile regression to determine how gender influences future revenue expectations. We control
for various characteristics such as education, experience, funding requests, family structure,
household responsibilities, business type and firm industry.
This study finds that gender does not influence firm survival; 71.3% of women and
68.7% of men who establish an operational firm were able to maintain a one-year surviving
business. We found no gender impact on future one-year annual revenue expectations. Initially
we find a negative correlation between female entrepreneurs and five-year annual revenue
expectations, but when we control for business type and firm industry, we find that the difference
is due to industry of the firm. This indicates that female entrepreneurs tend to choose industries
with lower expected revenues. Despite these similarities of initial firm success and future
financial expectations, we did find some notable differences in a few characteristics of the
entrepreneurs. Our results showed that depending on their business status, women spend between
8.1 and 12.5 more hours a week on housekeeping and childcare responsibilities on average than
their male counterparts. Women were significantly more likely than men to cite flexibility as an
important reason for creating their businesses; 84% of surviving female entrepreneurs stated they
value personal and family flexibility compared to around 60% for surviving male entrepreneurs.
5
In the initial survey, 43.9% of males indicated to have either already requested funding or
planned on requesting in the future, compared to 36.4% of women.
These findings indicate entrepreneurship may be a good option for women. While female
entrepreneurs share many of the characteristics of other females in the workforce,
entrepreneurship allows them to have managerial responsibilities and achieve success in a more
flexible environment. However, we find significant gender differences similar to those found in
previous glass ceiling literature. The women in this sample cite flexibility as an important
motivation for becoming an entrepreneur more than men and the women spend significantly
more time on domestic responsibilities. Though these characteristics do not impact initial
entrepreneurial success, it is possible they could lead to gender differences in longer term
performance for the female entrepreneurs since these gaps have been shown to hinder women’s
career progression in the corporate world.
The remaining structure of the paper is as follows. Section 2 explores previous literature
examining gender inequalities in the corporate and entrepreneurship labor markets. Section 3
describes the data set and the empirical framework for examining the impact of gender on initial
firm success and future financial expectations. Section 4 presents empirical evidence of gender
differences amongst the entrepreneurs and the effect of gender on our outcomes of interest.
Section 5 concludes the study and discusses opportunities for future research.
2. Gender Differences in Labor Markets
2.1 Traditional Labor Market
Research on factors contributing to the glass ceiling in the traditional labor market has
found that societal and personal barriers play a greater role in preventing women from rising to
high-paying, executive positions, than does blatant gender discrimination (Bertrand, 2017;
6
Goldin, 2014; Lyness & Grotto, 2018). Gender earnings gaps cannot be explained solely by
differences in occupations since women earn less than men both across and within the same
occupations. Women’s desire for more flexibility on the job, particularly regarding hours
worked, has been shown to negatively influence their ability to rise to top level positions – roles
which commonly require long hours and have little flexibility. This theory is supported by the
fact that we see a large deficit in wages for women and men in high-earnings occupations in
general. In these roles, hourly pay is often correlated with hours worked, thus rewarding those
who work longer, putting women at a disadvantage (Goldin, 2014).
Women’s desire for flexibility in their careers reflects gender stereotypes that are still
prevalent in relation to family roles and responsibilities. Alice Eagly, a pioneer of social role and
gender stereotype research, theorizes that the innate physical differences between men and
women have led to a division of roles in society based on gender and the development of gender
role stereotypes. These stereotypes impact how men and women conceive of their social roles,
such as a that of a parent and employee (Eagly & Wood, 2016). Since 1965, the average time
women spend caring for children has remained relatively stable despite an increase in women’s
labor participation over the same period. Although men now spend more time with their children
and take on more household labor than ever before, women still assume the majority of domestic
responsibilities (Pew Research Center, 2013). Women with children under 18 spend around 59%
of their total weekly work time on housework and childcare responsibilities. Comparatively, men
with children under 18 spend around 30% of their total weekly work time on these
responsibilities (Pew Research Center, 2013).
In addition, Gary Becker’s “A Theory of the Allocation of Time” finds that men and
women allocate their time and wealth to maximize family well-being (Becker, 1965). It is often
7
the case that women have the ability to carry out housework more efficiently than men due to
fact women are raised with different expectations and experience more frequent labor-force
discrimination (Blau, Ferber, & Winkler, 2010). Therefore, it is often expected that men focus on
financial support for the family and women focus more on emotional and personal support.
Career interruptions can also negatively affect hourly pay, which is often experienced when
women have children. Not only does this lead to exits from the workforce, but it also constrains
hours worked by the women, as they now have more personal responsibilities (Blau & Kahn,
2017). Family responsibilities have been found to negatively impact promotional opportunities
thereby disproportionally lowering the probability of advancement for women (Konrad &
Cannings, 1997; Ochsenfeld, 2012). These responsibilities are a double-edged sword; employers
are less likely to promote those who they know will have more personal obligations and women
may exert less effort in their paid work knowing they have to dedicate more time to family
responsibilities, all of which can hurt their career progression.
2.2 Entrepreneurial Market
Similar to corporate employees, research shows that decisions by entrepreneurs are often
impacted by personal and family responsibilities. Social roles influence women beginning with
their motivations for starting a business There are two main types of entrepreneurs – opportunity-
based (those that create businesses when they see a business opportunity), and necessity-based
(those who start a business due to need of employment) (Fairlie & Fossen, 2017). Previous
research has found opportunity-based entrepreneurs are more likely to be male, which suggests
women are more likely to choose entrepreneurship out of work necessity (Zwan et al., 2016).
Female entrepreneurs rank balancing a family as more of a significant factor in deciding to
pursue entrepreneurship compared to males – a preference that may lead them to participate
8
primarily in necessity-based entrepreneurship (Buttner & Moore, 1997; DeMartino & Barbato,
2003). Opportunity-based entrepreneurs tend to have more profitable businesses than necessity-
based entrepreneurs. (Calderon et al., 2017).
Another way family roles can impact female entrepreneurs can be inferred from
demographic data about entrepreneurs. The mean age of successful entrepreneurs in the U.S. at
the time of founding a company is around about 42 years-old. When looking at the average age
of founders of successful businesses that grow most quickly, the average age is even higher at 45
years-old (Azoulay, Jones, Kim, & Miranda, 2018). At this age, people are often married or have
children, At least 85% percent of women ages 40 to 49 in the United States are married (Kreider
& Ellis, 2011) and 86% percent of women ages 40 to 44 have given birth (Livingston, 2018).
Family situations can have a differential impact on women and men who are considering starting
a business.
Research also has identified differential treatment by investors unique to the
entrepreneurial market and unrelated to family roles. Discrimination in the general credit market
has often led to disadvantages for women. (Henderson, Herring, Horton, & Thomas, 2015). For
business loans specifically, women typically have a more difficult time securing funding than
men do, producing a phenomenon referred to as the funding gap (Brush et al., 2004; Carter &
Rosa, 1998). In terms of traditional credit loans, women who have larger businesses (in terms of
number of employees) that show strong financial performance, do not receive the same funding
as their male counterparts who show similar results (Eddleston, Ladge, Mitteness, &
Balachandra, 2016). In terms of equity investing, female entrepreneurs currently receive only 2%
of all venture capital funding (Zarya, 2018). Some theorize that investors have higher
expectations from women when hearing their pitches. Research found that when considering
9
investments in female-owned businesses, the nature of the questions posed by investors is often
more concerned with failure compared to being focused around growth when evaluating male-
owned businesses. This disadvantages women by making them less willing to take risks and may
result in a lower likelihood of attracting investment (Kanze et al., 2018). This disparity also
could influence the ability of female entrepreneurs to grow their business at the same rate as men
and therefore inhibit their capacity to reach the same level of performance (Alsos et al., 2006).
Previous literature indicates that women’s roles at home may influence career
advancement. This paper examines the early success of entrepreneurs and whether a gender
performance gap exists in these successful businesses. Drawing on the idea of the glass ceiling,
this paper examines whether gender-specific family roles, desires for flexibility, work
experience, business characteristics and growth aspirations impede women’s advancement in
entrepreneurship.
3. Methodology
This section will document high level information around the individuals in the data set
and our key outcomes of interest - early firm survival and future financial performance
expectations. Many independent variables related to characteristics of the individuals and firms
will be used to control for these outcomes, with gender as the baseline variable. We end the
section with a detailed explanation of the empirical strategy that will be used to analyze gender
differences in our outcomes.
3.1 Data Set
This study uses data from the Panel Study of Entrepreneurial Dynamics (PSED) that was
conducted at the University of Michigan through the Institute for Social Research. The PSED is a
longitudinal study that provides detailed information from a nationally representative sample of
10
those in the early stages of business start-up development. The study includes significant details
about the firm creation process and a wide range of information about the founders, while
tracking entrepreneurial initiatives from conception into the early years of an operational new
firm. The cohort used for this study is referred to as PSED II.
1
The individuals for PSED II were selected from 2005 to 2006. The PSED sample frame
was all individuals at least 18 years old residing throughout the mainland U.S. The sample was
identified through a random digit dialing methodology, with 64,622 households surveyed. To
qualify for the study, researchers started with a representative sample of households and through
phone interviews, screened individuals in order to meet four criteria: (1) they considered
themselves as involved in the firm creation process, (2) they engaged in some start-up activity in
the past 12 months, (3) they expected to own all or part of the new firm, and (4) the initiative had
not progressed to the point it may be considered an operating business. A final number of 1,214
active nascent (or aspiring) entrepreneurs was selected – 762 males and 452 females.
Table 1: Wave Counts
Total
Male
Percent
Males
Female
Percent
Females
Wave A / Year 0
1,214
761
62.7%
453
37.3%
Wave B / Year 1
972
597
61.4%
375
38.6%
Wave C / Year 2
746
458
61.4%
288
38.6%
Wave D / Year 3
527
335
63.6%
192
36.4%
Wave E / Year 4
435
274
63.0%
161
37.0%
Wave F / Year 5
375
243
64.8%
132
35.2%
Source: PSED II Survey. Note: Table 1 shows the counts and percentage distribution of
individuals in each wave of the survey broken by gender.
1
The PSED I cohort was sampled in from 1998 to 2000 during the height of the dot.com bubble, and this may have
skewed the sample given the unusual increase of entrepreneurial activity during that time. For the PSED II cohort, researchers
also developed a more reliable and efficient research protocol, as well as including a larger and more diverse group of
individuals.
11
After the initial screening, the individuals received follow-up phone interviews every 12
months for five years, resulting in a total of six interviews per individual, if applicable. The data
set refers to these six interviews as wave A, wave B, wave C, wave D, wave E and wave F. The
attrition from wave A through wave F reflects the loss of the individuals who did not wish to
participate or could not be located for more detailed interviews, as well as those who disengaged
from their business venture and were therefore no longer relevant to the study.
The average age of individuals in the wave A sample is about 44-years-old. Almost 79%
are Caucasian, followed by 12.3% African-American, 0.6% Asian or Asian-American and 8.4%
other races. Nearly all (94.6%) of the individuals were born in the United States. When looking
at females and males separately, the breakdowns of age and race are similar to that of the entire
sample.
Table 2: Age and Ethnicity of Nascent Entrepreneurs (N = 1214)
Total
Percent of
All
Percent of
Males
Percent of
Females
Mean Age
44
-
-
-
Median Age
44
-
-
-
Minimum Age
18
-
-
-
Maximum Age
83
-
-
-
Born in the U.S.
-
94.6%
94.7%
94.5%
White/Caucasian
-
78.8%
78.3%
79.6%
Black/African American
-
12.3%
11.7%
13.2%
Asian/Asian American
-
0.6%
0.8%
0.2%
Some other race
-
8.4%
9.2%
7.0%
Source: PSED II Survey. Note: Table 2 shows the age and ethnicity distribution of the
individuals in the sample.
12
About 37% of all the individuals have at least a college degree and about 39% have a
technical degree, vocational degree, community college degree or have attended some college.
This equates to about 76% of the sample having an education level above a high school degree.
According the U.S. Census, as of 2018, about 60.5% of Americans have an education level
above high school, indicating that this sample is more educated than that of the U.S. population
(U.S. Census Bureau, 2019). The women in the sample are slightly more educated; 81.2% of the
women have an education level above a high school degree compared to 72.6% for the men. This
is consistent with aggregate trends found in the U.S. population which show that more women
obtain secondary degrees than men (U.S. Census Bureau, 2019).
Table 3: Education and Employment of Nascent Entrepreneurs
Total
Percent of
All
Males
Percent of
Males
Females
Percent of
Females
Less Than High School
55
4.5%
41
5.4%
14
3.1%
High School Degree
238
19.6%
167
22.0%
71
15.7%
Some College, Technical, Vocational
or Community College Degree
471
38.9%
280
36.9%
191
42.2%
Bachelor’s Degree
255
21.0%
143
18.8%
112
24.7%
Graduate Training or Degree
193
15.9%
128
16.9%
65
14.3%
Currently working for pay
664
54.7%
442
58.1%
222
49.1%
Mean Years of Work Experience
22.8
-
22.3
-
23.2
-
Median Years of Work Experience
21
-
19
-
22
-
Mean Number of Previous Businesses
as Owner or Part-Owner
3
-
3.2
2.7
-
Median Number of Previous
Businesses as Owner or Part-Owner
1
-
1
-
1
-
Source: PSED II Survey. Note: Table 3 shows the counts and percentage distribution for education level, current work status,
work experience, and previous business ownership for the individuals in the sample.
13
At the time of interviewing, 54.7% of the sample were currently working for pay in some
capacity. Men were more likely to be working for pay than women by a difference of nine
percentage points. The average number of years of total work experience for the sample is about
22.8. The men have only about a year more of average work experience than the women.
As part of the survey’s criteria, we know that all the individuals are nascent entrepreneurs
at the initial screening. According to information from the survey, they all have had previous
experience as a business owner or part-owner. The average number of other businesses where the
individual had been either an owner or part-owner is three. There is no detailed information on
their previous business ventures, so it is unclear whether these were fully operational firms or
were in the pre-establishment phase. Overall, the individuals in the sample are educated, have a
significant amount of work experience and have previous entrepreneurial experience.
3.2 Outcomes of Interest
To test whether there are gender differences in entrepreneurial performance, this study
will use two different performance-level metrics as the outcomes of interest. The first metric is
firm survival rate. This tests whether the nascent entrepreneur is successful in creating an
operational firm and that the firm survives at least one year after creation. The second metric is
future annual revenue expectation. In this sample, self-reported information on revenues, costs
and market valuations are very limited due to lack of response from most of the individuals.
Therefore, this study chose to use expected annual revenue as way to measure potential financial
performance of the firms, due to their higher response rates.
Firm Survival. Just being able to survive and maintain operations can be seen as a form
of success for new businesses (Zolin, Stuetzer, & Watson, 2013). Many researchers have used
firm survival rate as a proxy for new business success (Cassar, 2014; Fried & Tauer, 2015; van
14
Gelderen, Thurik, & Bosma, 2005). The probability of failing is very high for the average new
business. The statistics on new business failure rates are varied, however the U.S. Small Business
Administration indicates that 50% fail after five years and 66% fail after ten years. (U.S. Small
Business Administration 2012).
In interview waves B to F, the PSED II researchers asked the individuals if their business
is considered either an operational firm, a non-operational active startup, or if operations have
been disengaged. An operational firm indicates that the business has become a fully operating
new firm, a non-operational active startup indicates that the individual is working on the firm,
but it has yet to become operational, and disengaged indicates that the individual has stopped all
business activity related to the firm. Individuals with disengaged operations are dropped from
future wave surveys.
To properly calculate the survival rate of the firm, we first created a variable that
indicates in what year the individual established their firm as operational. Then, based on the
creation year, we created a one-year survival variable. We did this by examining their creation
year and if they stated that they still had an operational firm in the following year. The trend of
businesses that went from nascent to established to surviving are shown in Figure 1. Of all the
nascent entrepreneurs, 18.7% (227/1214) were able to create an operational firm within the five
years of the survey. Of the entrepreneurs that established a firm, 69.6% (158/227) created a firm
that survived at least one year past creation. Though only 13% of all nascent entrepreneurs were
able to create a one-year surviving established firm, this disparity is due to low levels of
establishment, not survival.
15
Revenue Expectations. Regarding future revenue expectations, the individuals were
asked, “After this (new) business is operational, what is the total revenue expected in the first
twelve months of operation?” and “What annual revenue is expected when the business is in its
fifth year of operation?” We chose to analyze these outcomes only for those individuals that
created one-year surviving firms, because we wanted to compare the financial expectations of
females and males that were able to reach that initial level of success through one-year survival.
One possibility for the gender gap in revenues of entrepreneurs that was found by the U.S.
Census could be that female entrepreneurs have lower financial expectations than male
entrepreneurs. In terms of coding the variable, the individuals were asked this question in each
wave of the survey, so the most recent response prior to or in the year of creation of their firm
was used. The descriptive statistics in Table 5 for the one-year and five-year revenue
1,214
227 158
Nascent Entrepreneurs Established Entrepreneurs Surviving Entrepreneurs
Figure 1: Nascent, Established, and Surviving Entrepreneurs
Source: PSED II Survey. Note: Figure 1 shows the counts and distribution for nascent, established and surviving
entrepreneurs.
16
expectations show a large difference between the mean and median indicating that there are the
distribution is positively skewed.
Table 5: Annual Revenue Expectations for Established and Surviving Entrepreneurs
Mean
Std. Dev.
Median
Min
Max
Established
(N=227)
One-Year Annual
Revenue Expectations
$672,509
$6,910,489
$50,000
$60
$100,000,100
Five-Year Annual
Revenue Expectations
$1,152,017
$7,048,650
$120,000
$500
$100,000,300
Surviving
(N=158)
One-Year Annual
Revenue Expectations
$883,949
$8,238,895
$50,000
$600
$100,000,100
Five-Year Annual
Revenue Expectations
$1,283,010
$8,348,734
$100,000
$5,000
$100,000,300
Source: PSED II Survey. Note: Table 5 shows the descriptive statistics for one-year and five-year annual revenue expectations
for the entrepreneurs that operate an establish business for at least one year.
3.3 Independent Variables
Previous literature has found many factors can contribute to a firm’s ability to succeed
(Cassar, 2014; Fried & Tauer, 2015; Song, Podoynitsyna, Van Der Bij, & Halman, 2008; van
Gelderen et al., 2005). There has been a particular focus on individual level characteristics of the
entrepreneur versus firm level characteristics in predicting the initial success of a business. Some
of these individual characteristics include things such as education level, work experience,
previous entrepreneurial experience, and motivations. Firm level attributes such as funding and
industry have also been explored. This study will use several individual level and some firm
level characteristics as controls for the analysis to understand whether there are gender
differences in initial survival and financial performance expectations, independent of these
controls.
Gender. Gender will be used as the baseline independent variable to document gender
differences in firm one-year survival and future revenue expectations. In the initial wave A, the
individuals are asked to designate their gender; 62.7% of the entrepreneurs are male and 37.3%
17
are female. This breakdown is very similar to that of the U.S. Census data analysis which states
that female entrepreneurs make up 36% of business owners (McManus, 2017).
Education and Work Experience. In the initial wave A, the survey asks a variety of
questions related to the education and work experience of the individual. For education, the
respondent is asked to state their highest completed level. We combine the levels into five
categories, less than high school, high school degree, some college, college degree and graduate
training or degree. Regarding work experience, they are asked “How many years of full time,
paid work experience have you had?”, “How many years of work experience had you have in the
industry where this (new) business will compete?” and “How many other businesses have you
helped to start as an owner or part-owner?” Age is also used as a control variable related to
experience.
Motivations. In the initial wave A, the individuals are asked 14 questions related to their
motivation for starting the business. They answer all the questions on an ordinal scale of
importance: no extent, a little extent, some extent, a great extent, or a very great extent. Some of
these questions focus on financial motivations through questions like “To what extent is earning
a larger personal income important to you for establishing this business?” and “To what extent is
having a chance to build great wealth or a very high income important to you for establishing
this business?” Other questions focus more around lifestyle motivations through questions like
“To what extent is having greater flexibility of your personal and family life important to you for
establishing this business?” and “To what extent is having considerable freedom to adapt your
own approach to work important to you for establishing this business?” In order to analyze
importance vs. non-importance, we re-code this variable as binary. If the response is “a great” or
18
“a very great” extent, the answer is coded as high importance. The remaining responses are
coded as not high importance.
Family Responsibilities. In the initial wave A, the researcher asks questions related to
the family structure of the individual including marital status and children. The individuals are
asked about their marital status which included responses of “Married”, “Living with Partner”,
“Separated”, “Divorced”, “Widowed”, or “Single.” We create a categorical binary variable for
marital status; if the response is married or living with partner, the answer is coded as these
options and if the response is separated, divorced, widowed or single, the answer is coded as not
married or living with partner. For information on the whether the individual has children, they
ask, “How many children seventeen years and younger are living in your household?” We re-
code this to a binary variable as well, with one group having at least one child under the age of
17 and the other having no children under the age of 17. In terms of domestic responsibilities,
they ask the individuals “How many hours a week do you spend on housekeeping and childcare
activities?” and the responses are coded numerically.
Funding. The survey asks the individuals if they have requested or received outside
funding. The question “Have financial institutions or other people been asked for funds for this
new business, do you expect to ask for funds in the future, or is outside financial support not
relevant for this new business?” is used to determine need for funding. The question “Have you
received the first outside funding from financial institutions or other people for this new
business?” determines whether the individual has already received any funding. This second
question had a very low response rate however, so we used the request for funding question
instead as a proxy for funding needs. The individuals are asked this question in each wave of the
survey, so the most recent response prior to or in the year of creation of their firm is used.
19
Business Type. The individuals are also asked to state whether their business is
opportunity-based or necessity-based through the question “Are you involved in this (new)
business to take advantage of a business opportunity or because you have no better choices for
work?” If the individual responds “to take advantage of a business opportunity”, we code that
response as opportunity-based and if the response is “no better choice”, we code that response as
necessity-based. The individuals are asked this question in each wave of the survey, so the most
recent response prior to or in the year of creation of their firm is used.
Firm Industry. The individuals are also asked to designate the industry of their business
out of the following options: Retail Store; Restaurant Tavern; Bar or Nightclub; Customer or
Consumer Service; Health; Education or Social Service; Manufacturing; Construction;
Agriculture; Mining; Wholesale Distribution; Transportation; Utilities; Communications;
Finance; Insurance; Real Estate or Business Consulting Service. The individuals are asked this
question in each wave of the survey, so the most recent response prior to or in the year of
creation of their firm is used.
Funding requests, business type and firm industry variables could be estimated as
outcomes of interest –individual characteristics can influence these business-related decisions.
However, in this paper, they will be used as independent variables to see their effect on new
business success and performance expectations.
3.4 Empirical Strategy
In the first part of the data analysis, we will look at various firm and entrepreneur
characteristics broken out by gender to see if any significant differences exist. This includes
looking at gender differences among business type (opportunity-based or necessity-based), firm
industry, funding requests, motivation for starting the business and hours spent on childcare.
20
These are the major areas where previous literature has found gender differences for employees
both the in traditional and entrepreneurial markets.
Survival Logistic Regression. Due to the gender differences discussed in previous
research, we hypothesize that these gaps exist in this sample and may inhibit female-owned
businesses from reaching the same initial survival rates as male-owned businesses.
Hypothesis 1: Of the established new businesses, male-owned new businesses will have higher
one-year survival rates than female-owned new business.
To analyze the impact of gender on firm survival, a logistic regression is used. There will
be a baseline model for one-year firm survival of established firms that will analyze the effect of
being female on this outcome.
Where p is the probability of the established firm surviving one-year and x1 is as indicator
variable that takes on a value of one if the respondent is female and zero otherwise. Once results
for the baseline model have been estimated, we will then estimate four additional models. The
second model will add in controls for age, education, work experience and previous
entrepreneurial experience. The third model will add in controls for funding requests. The fourth
model will add in controls for family structure and domestic responsibilities. The final fifth
model will add in controls related to the firm type and industry.
In addition, OLS regressions will be estimated for each of these five models to test for
robustness of the results from the non-linear models. A non-linear model, such as the logistic
regression, is used to explain the relationship between one binary outcome variable and one or
more nominal, ordinal, interval or ratio-level independent variables. This was the chosen model
for the survival outcome since surviving is a binary variable – a business either survives one-year
(1)
21
or it does not. A linear model, such an OLS, assumes that the outcome variable is continuous,
which is violated in this case. However, OLS non-linear binary models are often estimated to
compare its coefficient results with the marginal effects from the logistic model, thus testing the
robustness of the logistic model.
Revenue Expectation Median Quantile Regression. We hypothesize that the gender
differences in entrepreneurs will not only impact initial survivability of these firms, but the
potential of reaching high financial performance as well. Research finds that women value
flexibility more than wealth when starting a business and are more likely to start necessity-based
businesses (Buttner & Moore, 1997; DeMartino & Barbato, 2003), all of which could influence
their financial growth expectations.
Hypothesis 2: Of the surviving new businesses, the male-owned businesses will have higher one-
year and five-year expected revenues than the female-owned businesses.
To analyze the impact of gender on one-year and five-year annual revenue expectations, a
quantile regression for the 50th percentile quantile (the median) will be used. Quantile
regressions do not assume a normal parametric distribution nor homogeneity of variance – two
assumptions required for OLS regressions. This decision was made because the distribution of
these two variables are heavily positively skewed. The baseline model will analyze the
effect of being female on one-year and five-year annual revenue expectations:
Where Q1 is one-year annual revenue expectations, Q5 is five-year annual revenue expectations
and x1 is gender with male as the base value. The second model add in controls for business type
and the third model adds in controls for firm industry.
(2)
(3)
22
4. Data Analysis
In the data analysis section, we first provide multiple descriptive statistics and then
estimate logistic and quantile regressions to understand gender differences among these
entrepreneurs and how they impact firm performance. The goal of this section is to gain a better
understanding of these gender differences and to analyze any patterns that might indicate that the
women experience barriers or inhibitors to entrepreneurial success.
3.1 Firm and Entrepreneur Characteristics
Business Type. The entrepreneurs in this sample are overwhelmingly creating
opportunity-based businesses – around 87% of the total firms. We found very small gender
differences among the nascent, established and surviving entrepreneurs as shown in Table 6.
Men are slightly more likely to have an opportunity-based business only within the established
cohort, not the nascent and surviving. However, none of these differences were statistically
significant at the 5% level when using the chi-square test.
Table 6: Business Type of Firms by Gender
Percent Total
Firms
Percent Male-
Owned Firms
Percent Female-
Owned Firms
Difference
(% Male - % Female)
Nascent
(N=1214)
Opportunity
86.6%
86.4%
87.2%
-0.8%
Established
(N=227)
Opportunity
85.7%
86.5%
84.2%
2.3%
Surviving
(N=158)
Opportunity
86.2%
84.7%
88.9%
-4.2%
Source: PSED II Survey. Note: Table 6 shows the distribution of the opportunity-based and necessity-based firms in the
nascent, established and surviving samples for the entire sample as well as gender subsets.
Industry. In addition to business type, we also look at industry distribution among the
firms. Of all the nascent entrepreneurs, the top three industries are Consumer Service (35%),
Retail Store (13%) and Business Consulting (7.9%).
23
We found nascent female entrepreneurs concentrated in fewer industries than male
entrepreneurs. Women are more likely to have a Communications, Consumer Service, Health
Service, Insurance, Real Estate, and Retail Store business. Men are more likely to have an
Agriculture, Business Consulting, Construction, Finance, Manufacturing, Mining, Restaurant,
Transportation and Wholesale Distribution business.
4.5%
7.9%
2.8%
6.8%
35.0%
1.5%
7.0%
0.7%
5.5%
0.2%
5.5%
3.6%
13.0%
1.7%
4.4%
0% 5% 10% 15% 20% 25% 30% 35% 40%
Agriculture
Business Consulting
Communications
Construction
Consumer Service
Finance
Health Service
Insurance
Manufacturing
Mining
Real Estate
Restauarnt
Retail Store
Transportaion
Wholesale Distribution
Figure 1: Industry Distribution for Nascent Entrepreneurs
Source: PSED II Survey. Note: Figure 1 shows the firm industry distribution among the nascent entrepreneur sample.
(N=1214).
24
The men and women are equally likely to have a Transportation business. However, only
Business Consulting Services, Construction, Health Service and Retail Store were found to have
statistically significant gender differences at the 5% level when using the chi-square test. There
were very little changes between the distributions of the industries for each gender as Table 7
showcases the industry gender differences among these three groups. The highlighted values
indicate when a statistically significant gender difference between share of industry was found.
Though gender differences were observed in the established and survival cohorts, the small
sample size may impact the statistical significance.
0% 5% 10% 15% 20% 25% 30% 35% 40%
Agriculture
Business Consulting
Communications
Construction
Consumer Service
Finance
Health Service
Insurance
Manufacturing
Mining
Real Estate
Restauarnt
Retail Store
Transportation
Wholesale Distribution
Figure 2: Industry Distribution for Nascent Entrepreneurs by Gender
Women Men
Source: PSED II Survey. Note: Figure 1 shows the firm industry distribution among the nascent entrepreneur sample broken
out by gender. (N=1214).
25
Table 7: Distribution of Firm Industries
Nascent (N=1214)
Established (N=227)
Surviving (N=158)
Male-
Owned
Firms
Female-
Owned
Firms
Male-
Owned
Firms
Female-
Owned
Firms
Male-
Owned
Firms
Female-
Owned
Firms
Agriculture
5.1%
3.5%
7.5%
2.4%
10.9%
3.5%
Business Consulting Services
9.2%
5.5%
11.6%
7.3%
14.9%
7.0%
Communications
2.6%
3.1%
2.0%
6.1%
2.0%
7.0%
Construction
9.5%
2.2%
8.8%
3.7%
7.9%
1.8%
Consumer Service
33.8%
36.9%
38.8%
30.5%
36.6%
36.8%
Finance
1.8%
0.9%
2.7%
1.2%
1.0%
1.8%
Health Service
4.4%
11.5%
5.4%
13.4%
5.9%
12.3%
Insurance
0.5%
1.1%
0.0%
3.7%
0.0%
3.5%
Manufacturing
6.2%
4.4%
6.1%
4.9%
5.0%
1.8%
Mining
0.3%
0.2%
0.0%
0.0%
0.0%
0.0%
Real Estate
4.9%
6.4%
5.4%
9.8%
4.0%
8.8%
Restaurant
4.0%
2.9%
1.4%
1.2%
2.0%
1.8%
Retail Store
10.3%
17.5%
5.4%
11.0%
5.0%
8.8%
Transportation
2.1%
0.9%
0.7%
1.2%
1.0%
1.8%
Wholesale Distribution
5.3%
2.9%
4.1%
3.7%
4.0%
3.5%
Source: PSED II Survey. Note: Table 7 shows the distribution of the firm industries in the nascent, established and
surviving samples, as well as gender subsets. The number that is shaded indicates the corresponding gender that has a
greater share of businesses within that industry.
Funding. Looking at individual funding requests, we find that women are more likely
than men to state that funding is not relevant to their business for all three cohorts – nascent,
established and surviving. Consequently, men are more likely to have either already requested
funding or plan on requesting it in the future. The established and surviving entrepreneurs exhibit
a wider gender gap than the nascent entrepreneurs. However, the only cohort that exhibits a
statistically significant gender difference at the 5% level through chi-square tests, is the nascent
entrepreneur, most likely due to the smaller sample sizes of the other cohorts.
26
Table 8: Funding Requests by Gender
Percent Total
Firms
Percent Male-
Owned Firms
Percent Female-
Owned Firms
Difference
(% Male - % Female)
Nascent
(N=1214)
No, Not Relevant
59.0%
56.2%
63.6%
-7.40%*
Not Yet, Will in Future
26.7%
28.2%
24.2%
4.00%*
Yes
14.4%
15.7%
12.2%
3.50%*
Established
(N=227)
No, Not Relevant
50.7%
45.8%
59.5%
-13.70%
Not Yet, Will in Future
17.9%
20.1%
13.9%
6.20%
Yes
31.4%
34.0%
26.6%
7.40%
Surviving
(N=158)
No, Not Relevant
52.5%
48.5%
59.6%
-11.10%
Not Yet, Will in Future
16.5%
17.8%
14.0%
3.80%
Yes
31.0%
33.7%
26.3%
7.40%
*p < 0.05
Source: PSED II Survey. Note: Table 8 shows the distribution of funding requests in the nascent, established and surviving
samples, as well as gender subsets.
Entrepreneur Motivation. Looking at the proportional differences of the importance of
various reasons for creating a firm in Table 9, we find that surviving female entrepreneurs are
more likely than surviving male entrepreneurs to cite the need for greater flexibility in relation to
their personal and family life as an important factor. This difference is significant at the 5% level
using the chi-square test. As the sample goes from nascent to established to surviving, the
importance of personal and family flexibility increases for women. Among the nascent and
established entrepreneurs, men value earning a higher income and building wealth as more
important than women. However, of the surviving entrepreneurs, female entrepreneurs rank
financial motivators higher in importance than the surviving male entrepreneurs.
27
Table 9: High Importance of Motivation for Creating Business by Gender
Percent
Female-
Owned
Firms
Percent
Male-
Owned
Firms
Difference
(% Female
– % Male)
Nascent
(N=1214)
Personal and family flexibility
71.50%
63.60%
7.90%
Freedom to adapt your own approach to work
75.30%
72.00%
3.30%
Earning a larger personal income
60.50%
62.50%
-2.00%
Opportunity to build great wealth or a very high income
33.10%
41.00%
-7.90%
Established
(N=227)
Personal and family flexibility
78.80%
65.30%
13.50%
Freedom to adapt your own approach to work
72.50%
68.00%
4.50%
Earning a larger personal income
60.00%
61.90%
-1.90%
Opportunity to build great wealth or a very high income
27.50%
35.40%
-7.90%
Surviving
(N=158)
Personal and family flexibility
84.2%
60.4%
23.8%*
Freedom to adapt your own approach to work
75.4%
66.3%
9.1%
Earning a larger personal income
56.1%
52.5%
3.6%
Opportunity to build great wealth or a very high income
28.1%
24.8%
3.3%
*p < 0.05
Source: PSED II Survey. Note: Table 9 shows the distribution of importance of motivations for starting a business in the
nascent, established and surviving samples, as well as gender subsets.
Exploring personal and family responsibilities a bit further, we break out average and
median weekly hours spent on housekeeping and childcare by gender for the nascent, established
and surviving samples in Table 10. A significant discrepancy between male and female
entrepreneurs is observed in each cohort – female entrepreneurs spend more time on
housekeeping and childcare than male entrepreneurs. As the sample goes from nascent to
established to surviving, the average and median number of hours increases for women.
28
However, for men, the average and median number of hours remains more stable. This is
consistent with the trend for women’s stated desire for more flexibility where the differences are
wider for established and surviving entrepreneurs than compared to the nascent entrepreneurs.
Table 10: Weekly Hours Spent on Housekeeping and Childcare by Gender
Female
Male
Difference
(Female - Male)
Nascent
(N=1214)
Mean
22.37
14.26
8.11*
Median
19
11
8
Established
(N=227)
Mean
23.98
12.90
12.90*
Median
21
11
10
Surviving
(N=158)
Mean
26.67
14.12
12.55*
Median
25
11
14
*p < 0.05
Source: PSED II Survey. Note: Table 10 shows the mean and median of number of
weekly hours spent on housekeeping and childcare for the nascent, established and
surviving samples broken out by gender.
3.1 Regression Analysis
Survival. From our initial analysis, we find that 18.7% of the nascent entrepreneurs are
able to create an operational firm at some point within the five years of the study. Breaking out
firm creation by gender, we see firm creation rates of 19.3% for male nascent entrepreneurs and
17.6% for female nascent entrepreneurs.
Table 11: Firm Creation Rate by Gender
Total Nascent Entrepreneurs (N=1214)
Established
Firms
Percent of
Established
Sample
Firms
Established
Male-
Owned
Firms
Percent of
Male-
Owned
Sample
Firms
Established
Female-
Owned
Firms
Percent
Female-
Owned
Sample
Firms
Difference
(% Male -
% Female)
Firm
Creation
227
18.7%
147
19.3%
80
17.6%
1.7%
Source: PSED II Survey. Note: Table 11 shows the total and gender counts and percent distribution of the nascent
entrepreneurs that are able to create an operational business at some point within the five years of the survey.
29
Analyzing one-year survival rates, we find that of the entrepreneurs that establish an operational
firm, 69.6% are able to survive at least one year. Breaking out one-year firm survival rates by
gender, we find 71.3% for female established entrepreneurs and 68.7% for male established
entrepreneurs.
Table 12: One-Year Survival Rate of Established Firms by Gender
Total Established Entrepreneurs (N=227)
Surviving
Firms
Percent
Surviving
Firms
Surviving
Male-
Owned
Firms
Percent
Surviving
Male-
Owned
Firms
Surviving
Female-
Owned
Firms
Percent
Surviving
Female-
Owned
Firms
Difference
(% Male -
% Female)
One-year
survival
158
69.6%
101
68.7%
57
71.3%
-2.6%
Source: PSED II Survey. Note: Table 12 shows the total and gender counts and percent distribution of the established
entrepreneurs that are able to create an operational business that survives at least one-year within the five years of the survey.
Though we found gender difference between the rates of the firm creation and survival
they are very minimal and not statistically significant at the 5% level. Looking at the fractions of
females for each of the three cohorts, we see that women consistently make up between 35.2%
and 37.3% of the total entrepreneurs within these groups.
62.7%
64.8% 63.9%
37.3%
35.2% 36.1%
0
100
200
300
400
500
600
700
800
Nascent Entrepreneurs Established Entrepreneurs Surviving Entrepreneurs
Counts
Figure 3: Gender Distribution of Nascent, Established and Surviving Entreprenuers
Men Women
Source: PSED Survey II. Note: Figure 3 shows the gender distribution within each of the three samples – nascent,
established and surviving.
30
Five logistic regressions were estimated for the one-year survival outcome of the
established firms. The first model estimates the probability of the individual creating a firm that
remains in operation for a year, controlling for gender. This tests our first hypothesis that of
those who establish an operational firm, the probability of creating a one-year surviving firm is
higher for men than for women. The second model adds in controls related to the individual’s
experience including age, education (base value is less than high school), total years of work
experience, total years of relevant industry experience and previous businesses created. The third
model adds in controls related to requests for funding (base value is funding request is “not
relevant”). The fourth model adds in controls related to family structure including number of
hours spent on weekly housekeeping and childcare, marital status, and whether the owner has
children under the age of 17-years-old. The final fifth model adds in controls for business type
(base value is necessity-based) and firm industry (base value is Agriculture).
Findings in Table 13 indicate that gender is not a significant factor in determining the
probability of one-year survival of a new firm. When adding in controls for funding requests in
the third model and for family structure in the fourth model, we find a statistically significant
positive impact of age of the entrepreneur. When we add in firm information in our final model,
we find a statistically significant positive impact of age and a negative impact of years of total
work experience. However, when looking at the marginal effect analysis, we find a very small
effect size for both variables. An increase of age increases the probability of a firm surviving one
year by 0.6 percentage points and an increase in number of years of work experience decreases
the probability by 0.4 percentage points. These results do not support our hypothesis that male
entrepreneurs survive at higher rates than female entrepreneurs.
31
Table 13: One-Year Firm Survival Logistic Regressions
Dependent Variable: One-Year of Firm Survival (N=227)
Coefficient
SE
Odds
Ratio
95% OR
Lower CI
95% OR
Upper CI
Marginal
Effect
Model 1
Female
0.082
0.380
1.085
0.522
2.340
0.013
Model 2
Female
0.117
0.403
1.124
0.516
2.537
0.004
Age
0.029
0.017
1.029
0.996
1.064
0.007
Education Level – High School
1.673
1.460
5.327
0.190
96.975
0.043
Education Level – Some College
0.909
1.294
2.482
0.106
29.483
0.222
Education Level- - College
0.034
1.298
1.034
0.044
12.364
0.149
Education Level - Graduate
0.224
1.311
1.251
0.052
15.278
-0.003
Years Total Work Experience
-0.021
0.014
0.979
0.953
1.007
0.017
Years Relevant Work Experience
0.019
0.015
1.02
0.992
1.051
0.003
Number of Previous Businesses
Created
-0.017
0.056
0.983
0.883
1.103
-0.003
Model 3
Female
0.020
0.431
1.020
0.443
2.432
0.003
Age
0.038
0.018
1.039
1.003
1.077
0.005*
Education Level – High School
2.344
1.657
10.424
0.301
386.733
0.241
Education Level – Some College
1.032
1.323
2.807
0.116
35.133
0.153
Education Level- - College
-0.071
1.320
0.932
0.039
11.535
-0.014
Education Level - Graduate
0.226
1.335
1.254
0.051
15.926
0.041
Years Total Work Experience
-0.024
0.015
0.976
0.948
1.006
-0.003
Years Relevant Work Experience
0.028
0.016
1.029
0.998
1.064
0.004
Number of Previous Businesses
Created
-0.049
0.060
0.953
0.849
1.075
-0.007
Funding Requested - Yes
0.136
0.600
1.145
0.372
4.050
0.017
Funding Requested – Not Yet,
Will in Future
-0.151
0.450
0.860
0.356
2.108
-0.021
Model 4
Female
-0.322
0.483
0.724
0.279
1.889
-0.042
Age
0.051
0.020
1.052
1.012
1.096
0.006*
Education Level – High School
2.273
1.729
9.713
0.251
403.916
0.193
Education Level – Some College
0.781
1.393
2.184
0.082
30.997
0.100
Education Level- - College
-0.478
1.397
0.620
0.023
8.757
-0.082
32
Table 13 Continued
Coefficient
SE
Odds
Ratio
95% OR
Lower CI
95% OR
Upper CI
Marginal
Effect
Education Level - Graduate
0.003
1.409
1.003
0.036
14.346
0.000
Years Total Work Experience
-0.028
0.016
0.972
0.942
1.003
-0.004
Years Relevant Work Experience
0.031
0.017
1.031
0.999
1.068
0.004
Number of Previous Businesses
Created
-0.067
0.064
0.935
0.824
1.064
-0.009
Funding Requested – Yes
0.545
0.651
1.725
0.512
6.788
0.062
Funding Requested – Not Yet,
Will in Future
-0.065
0.482
0.937
0.366
2.454
-0.009
Weekly Hours of Housekeeping
and Childcare
0.038
0.020
1.039
1.000
1.083
0.005
Married or Living with Partner
0.658
0.492
1.931
0.733
5.121
0.084
Has Children Under 17-years-old
0.024
0.598
1.024
0.318
3.387
0.003
Model 5
Female
-0.803
0.567
0.448
0.143
1.355
-0.088
Age
0.059
0.023
1.060
1.014
1.111
0.006*
Education Level – High School
3.891
2.021
48.968
0.897
4212.969
0.378
Education Level – Some College
2.026
1.639
7.587
0.220
265.042
0.276
Education Level- - College
0.660
1.655
1.934
0.054
68.876
0.105
Education Level - Graduate
0.972
1.663
2.643
0.073
95.196
0.151
Years Total Work Experience
-0.034
0.017
0.967
0.934
0.999
-0.004*
Years Relevant Work Experience
0.024
0.018
1.025
0.990
1.064
0.003
Number of Previous Businesses
Created
-0.115
0.076
0.891
0.766
1.035
-0.013
Funding Requested – Yes
-0.062
0.563
1.947
0.456
10.152
0.066
Funding Requested – Not Yet,
Will in Future
0.666
0.780
0.940
0.312
2.885
-0.007
Weekly Hours of Housekeeping
and Childcare
0.043
0.022
1.044
1.001
1.094
0.005
Married or Living with Partner
0.711
0.565
2.037
0.674
6.303
0.078
Has Children Under 17-years-old
0.203
0.704
1.226
0.309
5.026
0.022
Opportunity Business
-1.206
0.839
0.299
0.048
1.371
-0.108
Industry – Business Consulting
Services
-15.467
1707.369
0.000
0.00
1.60E+29
-0.064
Industry – Communications
0.340
3281.243
1.405
0.00
1.16E+26
0.000
Industry – Construction
-18.408
1707.369
0.000
0.00
1.04E+22
-0.426
Industry – Consumer Service
-16.944
1707.369
0.000
0.00
1.13E+27
-0.193
33
Table 13 Continued
Coefficient
SE
Odds
Ratio
95% OR
Lower CI
95% OR
Upper CI
Marginal
Effect
Industry – Finance
-18.505
1707.369
0.000
0.00
5.10E+25
-0.444
Industry – Health Service
-16.561
1707.369
0.000
0.00
8.63E+25
-0.149
Industry – Insurance
0.825
4505.674
2.282
0.00
9.84E+58
0.000
Industry – Manufacturing
-16.775
1707.370
0.000
0.00
6.10E+26
-0.173
Industry – Real Estate
-17.268
1707.369
0.000
0.00
7.01E+28
-0.237
Industry – Restaurant
-0.100
3676.681
0.905
0.00
7.35E+49
0.000
Industry – Retail
-17.058
1707.369
0.000
0.00
5.31E+27
-0.208
Industry - Transportation
-1.543
4389.688
0.214
0.00
1.37E-285
0.000
Industry – Wholesale Distribution
-17.810
1707.369
0.000
0.00
6.64E+24
-0.321
*p < 0.05
Source: PSED II Survey. Note: Table 13 shows results of the logistic regression of the probability of established entrepreneurs
creating an operational business that survives at least one-year within the five years of the survey.
We run an OLS linear-probability regression to test for robustness of the logistic
marginal effect estimates. In these results, age is seen as having a positive impact on the
probability of one-year firm survival in Models 3, 4, and 5. The effect of an increase in age in
Model 3 is correlated to a corresponding increase of 0.05 percentage points for the probability of
the firm surviving one year and an increase of 0.7 percentage points in Models 4 and 5.
However, years of work experience does not have a significantly negative impact in Model 5 as
it does in the logistic regression. Also, in this OLS model, the Construction and Finance industry
do have negative effects on the probability of firm one-year survival. A Construction firm
decreases the probability of a firm surviving one year by 34.3 percentage points and a Finance
firm decreases the probability by 47.3 percentage points. Within the established entrepreneurs,
the sample used for this regression, men have a greater share of Construction and Finance firms,
though the difference was not found to be statistically significant at the 5% level.
34
Table 14: One-Year Firm Survival OLS Regressions
Dependent Variable: One-Year of Firm Survival (N=227)
Model 1
Model 2
Model 3
Model 4
Model 5
Female
0.013
0.017
0.005
-0.039
-0.077
[0.059]
[0.061]
[0.059]
[0.063]
[0.067]
Age
0.005
0.005*
0.007*
0.007*
[0.003]
[0.003]
[0.003]
[0.003]
Education Level – High School
0.236
0.264
0.220
0.457
[0.246]
[0.238]
[0.236]
[0.289]
Education Level – Some College
0.157
0.166
0.129
0.341
[0.237]
[0.228]
[0.226]
[0.282]
Education Level – College
0.015
-0.001
-0.041
0.199
[0.240]
[0.230]
[0.229]
[0.285]
Education Level – Graduate
0.056
0.060
0.031
0.199
[0.240]
[0.231]
[0.229]
[0.285]
Years Total Work Experience
-0.003
-0.004
-0.004
-0.004
[0.002]
[0.002]
[0.002]
[0.002]
Years Relevant Work Experience
0.003
0.004
0.004
0.003
[0.002]
[0.002]
[0.002]
[0.002]
Number of Previous Businesses Created
-0.003
-0.007
-0.009
-0.011
[0.009]
[0.009]
[0.009]
[0.009]
Funding Requested – Not Yet, Will in Future
0.026
0.046
0.055
[0.082]
[0.082]
[0.088]
Funding Requested – Yes
-0.021
-0.022
-0.006
[0.065]
[0.067]
[0.070]
Weekly Hours of Housekeeping and Childcare
0.004
0.004
[0.002]
[0.003]
Married or Living with Partner
0.089
0.096
[0.067]
[0.070]
Has Children Under 17-years-old
0.028
0.032
[0.083]
[0.089]
Opportunity Business
-0.117
[0.089]
Industry - Business Consulting Services
0.014
[0.144]
Industry - Communications
0.090
[0.209]
Industry - Construction
-0.343*
[0.160]
Industry - Consumer Service
-0.138
[0.125]
35
Table 14 Continued
Model 1
Model 2
Model 3
Model 4
Model 5
Industry - Finance
-0.473*
[0.207]
Industry - Health Service
-0.115
[0.154]
Industry - Insurance
0.129
[0.305]
Industry - Manufacturing
-0.103
[0.188]
Industry - Real Estate
-0.233
[0.156]
Industry - Restaurant
-0.042
[0.248]
Industry - Retail Store
-0.144
[0.157]
Industry - Transportation
-0.136
[0.305]
Industry - Wholesale Distribution
-0.303
[0.174
R2
0.000
0.056
0.085
0.124
0.215
*p < 0.05
Source: PSED II Survey. Standard Errors are in brackets. Note: Table 14 shows results of the OLS linear-probability
regression of the probability of established entrepreneurs creating an operational business that survives at least one-year within
the five years of the survey.
Revenue Regressions. Looking at the proportional differences between future annual
revenue expectations broken out by gender in Table 15, a noticeable difference between males
and females is observed for the surviving entrepreneurs– men have higher mean and median
annual revenue expectations. However, the mean differences are not statistically significant at
the 5% level when using a t-test.
Due to the skewness of these variables, three median quantile regressions are estimated
for annual revenue expectations of the individuals that create a firm that survives at least one
year. The first model estimates the level of one-year and five-year annual revenue expectations
when controlling for gender. This tests our second hypothesis that an increase in expected annual
revenues of one-year surviving firms is correlated with being male. The second model adds in
36
controls for the business type (base value is with necessity-based) and the third model adds in
controls for the industry (base value is Agriculture).
Table 15: Annual Revenue Growth Expectations of One-Year Surviving Firms by Gender
One-Year Annual Revenue Expectations
Five-Year Annual Revenue Expectations
Surviving
Men-Owned
Firms
Surviving
Women-
Owned
Firms
Difference
(Men –
Women)
Surviving
Men-Owned
Firms
Surviving
Women-
Owned
Firms
Difference
(Men –
Women)
Surviving
(N=158)
Mean
$1,349,589.00
$88,129.09
$1,261,459.91
$1,803,570.00
$402,790.90
$1,400,779.10
Median
$60,000.00
$35,000.00
$25,000.00
$180,000.00
$75,000.00
$105,000.00
Source: PSED II Survey. Note: Table 15 shows the descriptive statistics for one-year and five-year annual revenue
expectations for the entrepreneurs that operate an establish business for at least one year, broken out by gender.
Findings in Table 16 indicate that gender is a not significant factor in determining the
one-year annual revenue expectations of the firm but is significant in determining the five-year
annual revenue expectations. The gender coefficient for the five-year annual revenue outcome is
negative indicating that being female is correlated to lower five-year annual revenue
expectations. When controls for firm business type are added in, neither gender nor business type
are significant for both outcomes. However, when the final controls for firm industry are
included, significance is found for the Finance, Restaurant and Transportation industries,
underscoring the main driver of these revenue expectations is not gender, but in fact firm
industry. The coefficients on these select industries are positive indicating that they are
correlated to higher one-year and five-year annual revenue expectations. Though these results do
not support our second hypothesis that male entrepreneurs have higher future revenue
expectations than female entrepreneurs, on average, women choose businesses in industries with
lower expected revenues.
37
Table 16: Annual Revenue Expectations of One-Year Surviving Firms Median Quantile Regressions
Dependent Variable: Annual Revenue Expectations (N=158)
One-Year Annual Revenue
Expectations
Five-Year Annual Revenue
Expectations
Model 1
Model 2
Model 3
Model 1
Model 2
Model 3
Female
-25,000
-24,000
-15,000
-105,000**
-99,000
-35,000
[13,300]
[14,977]
[18,924]
[39,147]
[53,646]
[58,694]
Opportunity Business
26,000
9,000
66,000
15,000
[17,651]
[19,633]
[61,810]
[77,683]
Industry - Business Consulting
0
100,000
[62,103]
[171,621]
Industry - Communications
15,000
135,000
[74,475]
[375,193]
Industry - Construction
0
-50,000
[63,385]
[172,057]
Industry - Consumer Service
-71,000
-240,000
[58,605]
[130,214]
Industry - Finance
400,000**
1,700,000**
[89,892]
[272,743]
Industry - Health Service
-58,000
-215,000
[62,060]
[145,049]
Industry - Insurance
-35,000
-115,000
[78,032]
[216,551]
Industry - Manufacturing
200,000
0
[133,483]
[307,791]
Industry - Real Estate
-15,000
-190,000
[66,222]
[151,011]
Industry - Restaurant
324,000**
750,000*
[101,233]
[367,552]
Industry - Retail Store
5,000
85,000
[79,015]
[205,346]
Industry - Transportation
90,0000**
5,700,000**
[89,892]
[272,743]
Industry – Wholesale Dist.
-60,000
-170,000
[62,990]
[160,119]
*p < 0.05 **p < 0.01
Source: PSED II Survey. Standard Errors are in brackets. Note: Table 16 shows results of the OLS regression of the one-year
and five-year annual revenue expectations of one-year surviving entrepreneurs.
38
5. Discussion
5.1 Overview of Findings
This paper set out to examine whether the glass ceiling patterns observed in the
traditional corporate workplace, are also observed in the entrepreneurial workplace. We may
assume that women who choose to become business owners to bypass the traditional glass
ceiling constraints, yet U.S. Census data observes a gap between share of female and male
entrepreneurs and an even larger gap between revenues generated by female-owned businesses
versus revenues generated by male-owned businesses. Though research on the gender pay gap
and glass ceiling in the traditional corporate world spans decades, empirical research analyzing
this phenomenon in the entrepreneurial world is less common.
This study finds that nascent female and male entrepreneurs have similar abilities to
create one-year surviving firms – about 71.3% of women and 68.7% of men who establish an
operational firm were able to survive at least one year. When we control for education,
experience, funding request, family structure, and firm characteristics, we did not find that
gender was correlated with firm survival. Though we found statistically significance in two
variables, age and years of work experience, the marginal effect was practically zero.
When looking at gender as a factor of future one-year revenue expectations, this study
finds no evidence of correlation. When looking at the correlation between gender and five-year
annual revenue expectations, we do find a statistically significant negative impact of being
female. However, when we control for business type and firm industry, we find that the
difference is due to industry of the firm, meaning women are more likely to be business owners
in industries with lower expected revenues.
39
Despite these similarities, we did find some notable differences in a few characteristics of
the entrepreneurs. Of the total nascent entrepreneurs in the initial survey, 63.6% of women
indicated that funding was not relevant to their business and had no plans for funding requests,
compared to 56.2% of men. Our results showed that of the surviving entrepreneurs, women
spend about 12.5 more hours a week on housekeeping and childcare on average than their male
counterparts, a statistically significant difference and similar results found by the Pew Research
Center. This study finds similar results found by Butter & Moore and DeMartino & Barbato, in
that women were significantly more likely than men to cite flexibility as an important reason for
creating their businesses; 84% of surviving female entrepreneurs stated they value personal and
family flexibility compared to around 60% for surviving male entrepreneurs. We found that
surviving female entrepreneurs were slightly more likely than surviving male entrepreneurs to
own an opportunity-based business; 88.9% of the surviving firms owned by women are
opportunity-based compared to 84.7% of the surviving firms owned by men. This result is
directionally different to the research of Calderon et al and Zwan et al., but in contrast is not
statistically significant, most likely due to the small sample size.
5.2 Interpretation of Findings
The U.S. has made great strides in increasing employment opportunities for women in the
corporate and entrepreneurial worlds. Almost half of employees of the S&P 500 companies are
women (Catalyst, 2018) and about 36% of new businesses created in the United States are
female-owned (McManus, 2017). However, we see a gender gap persist in both these markets in
terms of high level of performance – executive positions in the traditional corporate market and
high business revenues in the entrepreneurial market. Research around the gender gap in each of
these markets emphasizes the impact of family gender roles for both traditional corporate
40
employees and entrepreneurs. Despite an increase in male participation, women still take on a
majority of the household and childcare responsibilities. Though this work is technically unpaid,
it is no less difficult or time consuming and can hinder their ability to reach high levels of career
success. This disparity has led women to value flexibility as a key aspect of their career
The main takeaway from this study is that by looking at initial levels of success and
financial performance expectations among a group of aspiring entrepreneurs, women and men
are equally capable and ambitious within industry. One could argue that when the corporate
male-driven hierarchies are removed from women’s ability to reach managerial positions (i.e.
business owner), they find greater success. Perhaps entrepreneurship is a good solution for
women who want to reach those leadership positions, but also want to be able to balance their
personal and family responsibilities. However, we observe gender differences that are similar to
the differences found in research related to the corporate glass ceiling including women’s desire
for flexibility and time spent on family related responsibilities. The choice of women to pursue
businesses in industries with lower future revenue expectation may be linked to these high levels
of domestic responsibilities and flexibility preferences.
Though this study does not have data regarding longer term financial performance, it is
interesting to note that gender differences that have been linked to hindering longer term
performance of women in the corporate world have also been found in this study. Often, women
and men enter the labor force on similar levels, but their performance diverges overtime
(Bertrand, Goldin, & Katz, 2010; Goldin, 2014). These gender differences did not impact the
initial success and future revenue aspirations of these entrepreneurs, but it is possible that these
differences could impact the future performance of the female entrepreneurs differently than the
males.
41
5.3 Limitations and Future Research
There are some methodological limitations to this study. A major limitation to this data
set is the small sample size. Though the entire nascent cohort is 1,214 individuals, we primarily
look at established and surviving entrepreneurs which limits the subsequent samples to 227 and
158, respectively. We find some gender differences in variables yet were not able to find
statistical significance possibly due to the small sample size. Therefore, our results where we
observe a gender difference but did not find significance are inconclusive.
This sample also lacks robust revenue and other financial metrics of the established firms.
Since the focus of the data collection of this survey was on the aspiring entrepreneur and not the
established, this type of information is limited and therefore cannot be analyzed appropriately.
For the individuals that do report revenue and other performance metrics such as market value,
the numbers are self-reported, which also can create a response bias leading to unreliable data.
Another limitation of this study was the focus on just one-year of firm survival as one of
the outcomes of interest. Research has stated that the first five years of a business are vital in
determining their future success (Koulopoulos, 2015). However, due to the short timeframe of
this survey, we are not able to track most of the established firms to their five-year mark. Though
one-year survival is a good way to analyze initial success, one could argue that it is not enough
time to determine the longevity of a business. An opportunity for future research would be to use
a data set that focuses on already established entrepreneurs and then following their performance
for a few years to obtain more robust, longer term performance and financial metrics.
Also, due to the extensiveness of this data set, there are multiple ways that one could
compare the females and males in the sample. This study chose to focus on areas that have been
referenced in previous glass ceiling literature, however future research studies using this data set
42
could look at the impact of additional characteristics on firm survival and revenue expectations.
As more data becomes available around entrepreneurship, we look forward to seeing more
research in this topic.
43
References
Alsos, G. A., Isaksen, E. J., & Ljunggren, E. (2006). New Venture Financing and Subsequent
Business Growth in Men– and Women–Led Businesses. Entrepreneurship Theory and
Practice, 30(5), 667–686. https://doi.org/10.1111/j.1540-6520.2006.00141.x
Alsos, G. A., & Ljunggren, E. (2017). The Role of Gender in Entrepreneur–Investor
Relationships: A Signaling Theory Approach. Entrepreneurship Theory and Practice,
41(4), 567–590. https://doi.org/10.1111/etap.12226
Azoulay, P., Jones, B., Kim, J. D., & Miranda, J. (2018). Age and High-Growth
Entrepreneurship (No. w24489). https://doi.org/10.3386/w24489
Becker, G. S. (1965). A Theory of the Allocation of Time. Economic Journal, 75(299), 493–517.
Bertrand, M. (2017). The Glass Ceiling. Becker Friedman Instittue for Research in Economics,
Working Paper No. 2018-38. Retrieved from https://papers.ssrn.com/abstract=3191467
Bertrand, M., Goldin, C., & Katz, L. F. (2010). Dynamics of the Gender Gap for Young
Professionals in the Financial and Corporate Sectors. American Economic Journal:
Applied Economics, 2(3), 228–255.
Blau, F. D., Ferber, M. A., & Winkler, A. E. (2010). The Economics of Women, Men, and Work
(6th ed.). Pearson Education, Inc.
Blau, F. D., & Kahn, L. M. (2017). The Gender Wage Gap: Extent, Trends, and Explanations.
Journal of Economic Literature, 55(3), 789–865. https://doi.org/10.1257/jel.20160995
Brush, C. G., Carter, N. M., Gatewood, E. J., Greene, P. G., & Hart, M. (2004). Gatekeepers of
Venture Growth: A Diana Project Report on the Role and Participation of Women in the
Venture Capital Industry (SSRN Scholarly Paper No. ID 1260385). Retrieved from
Social Science Research Network website: https://papers.ssrn.com/abstract=1260385
44
Buttner, E. H., & Moore, D. P. (1997). Women’s Organizational Exodus to Entrepreneurship:
Self-Reported Motivations and Correlates with Success. Journal of Small Business
Management, 15.
Calderon, G., Iacovone, L., & Juarez, L. (2017). Opportunity versus Necessity: Understanding
the Heterogeneity of Female Micro-Entrepreneurs. The World Bank Economic Review,
30, S86–S96. https://doi.org/10.1093/wber/lhw010
Carter, S., & Rosa, R. (1998). The financing of male– and female–owned businesses.
Entrepreneurship & Regional Development, 10(3), 225–242.
https://doi.org/10.1080/08985629800000013
Cassar, G. (2014). Industry and startup experience on entrepreneur forecast performance in new
firms. Journal of Business Venturing, 29(1), 137–151.
https://doi.org/10.1016/j.jbusvent.2012.10.002
Catalyst. (2018, October 3). Pyramid: Women in S&P 500 Companies [Text]. Retrieved October
29, 2018, from https://www.catalyst.org/knowledge/women-sp-500-companies
DeMartino, R., & Barbato, R. (2003). Differences Between Women and Men MBA
Entrepreneurs: Exploring Family Flexibility and Wealth Creation as Career Motivators.
Journal of Business Venturing, 18(6), 815–832. https://doi.org/10.1016/S0883-
9026(03)00003-X
Eagly, A. H., & Wood, W. (2016). Social Role Theory. In The SAGE Encyclopedia of Theory in
Psychology (Vols. 1–2, pp. 895–898). https://doi.org/10.4135/9781483346274
Eddleston, K. A., Ladge, J. J., Mitteness, C., & Balachandra, L. (2016). Do you See what I See?
Signaling Effects of Gender and Firm Characteristics on Financing Entrepreneurial
45
Ventures. Entrepreneurship Theory and Practice, 40(3), 489–514.
https://doi.org/10.1111/etap.12117
Eisenmann, T. (2013, January 10). Entrepreneurship: A Working Definition. Retrieved June 1,
2019, from Harvard Business Review website: https://hbr.org/2013/01/what-is-
entrepreneurship
Fairlie, R. W., & Fossen, F. M. (2017). Opportunity versus Necessity Entrepreneurship: Two
Components of Business Creation. Stanford Institue for Economic Policy Research, 49.
Fried, H. O., & Tauer, L. W. (2015). An entrepreneur performance index. Journal of
Productivity Analysis, 44(1), 69–77. https://doi.org/10.1007/s11123-015-0436-0
Glass Ceiling Commission, U. S. (1995). Glass Ceiling Commission - A Solid Investment :
Making Full Use of the Nation’s Human Capital. Federal Publications. Retrieved from
https://digitalcommons.ilr.cornell.edu/key_workplace/120
Goldin, C. (2014). A Grand Gender Convergence: Its Last Chapter. The American Economic
Review, 104(4), 1091–1119.
Henderson, L., Herring, C., Horton, H. D., & Thomas, M. (2015). Credit Where Credit is Due?:
Race, Gender, and Discrimination in the Credit Scores of Business Startups - Loren
Henderson, Cedric Herring, Hayward Derrick Horton, Melvin Thomas, 2015. The Review
of Black Political Economy, 42(4). Retrieved from
https://journals.sagepub.com/doi/abs/10.1007/s12114-015-9215-
4#articleCitationDownloadContainer
Kanze, D., Huang, L., Conley, M. A., & Higgins, E. T. (2018). We Ask Men to Win and Women
Not to Lose: Closing the Gender Gap in Startup Funding. Academy of Management
Journal, 61(2), 586–614. https://doi.org/10.5465/amj.2016.1215
46
Konrad, A. M., & Cannings, K. (1997). The Effects of Gender Role Congruence and Statistical
Discrimination on Managerial Advancement. Human Relations, 50(10). Retrieved from
https://doi.org/10.1177/001872679705001006
Koulopoulos, T. (2015, October 21). 5 of the Most Surprising Statistics About Startups.
Retrieved December 13, 2018, from Inc.com website: https://www.inc.com/thomas-
koulopoulos/5-of-the-most-surprising-statistics-about-start-ups.html
Kreider, R. M., & Ellis, R. (2011). Number, Timing, and Duration of Marriages and Divorces:
2009. U.S. Census Bureau Current Population Reports, P70(125), 24.
Livingston, G. (2018, January 18). U.S. Women More Likely to Have Children Than a Decade
Ago. Retrieved January 13, 2019, from Pew Research Center website:
http://www.pewsocialtrends.org/2018/01/18/theyre-waiting-longer-but-u-s-women-today-
more-likely-to-have-children-than-a-decade-ago/
Lyness, K. S., & Grotto, A. R. (2018). Women and Leadership in the United States: Are We
Closing the Gender Gap? Annual Review of Organizational Psychology and
Organizational Behavior, 5(1), 227–265. https://doi.org/10.1146/annurev-orgpsych-
032117-104739
McManus, M. J. (2017). Women’s Business Ownership: Data from the 2012 Survey of Business
Owners. (No. 13). Retrieved from U.S. Small Business Administration website:
https://www.sba.gov/sites/default/files/advocacy/Womens-Business-Ownership-in-the-
US.pdf
National Women’s Business Council. (2012). Gender Differences in U.S. Business Fact Sheet.
Retrieved from https://s3.amazonaws.com/nwbc-prod.sba.fun/wp-
content/uploads/2012/01/05044441/fact-sheet-gender-differences-in-us-business.pdf
47
Nobel, C. (2011, March 7). Why Companies Fail—and How Their Founders Can Bounce Back.
Retrieved December 13, 2018, from HBS Working Knowledge website:
http://hbswk.hbs.edu/item/why-companies-failand-how-their-founders-can-bounce-back
Ochsenfeld, F. (2012). Glass ceiling or golden cage: Is the rise of women in first management
positions due to company discrimination or family responsibilities? Cologne Journal of
Sociology and Social Psychology, 64(3), 507–534.
Pew Research Center. (2013, March 14). Chapter 5: Americans’ Time at Paid Work, Housework,
Child Care, 1965 to 2011. Retrieved October 26, 2018, from Pew Research Center
website: http://www.pewsocialtrends.org/2013/03/14/chapter-5-americans-time-at-paid-
work-housework-child-care-1965-to-2011/
Song, M., Podoynitsyna, K., Van Der Bij, H., & Halman, J. I. M. (2008). Success Factors in New
Ventures: A Meta‐analysis. The Journal of Product Innovation Management, 25(1), 7–
27.
United States Department of Labor. (2016). Women in the Labor Force. Retrieved October 29,
2018, from United States Department of Labor website:
https://www.dol.gov/wb/stats/NEWSTATS/facts/women_lf.htm#one
U.S. Census Bureau. (2019, February 21). Educational Attainment in the United States: 2018.
Retrieved May 3, 2019, from https://www.census.gov/data/tables/2018/demo/education-
attainment/cps-detailed-tables.html
U.S. Department of Commerce. (2016, September). Characteristics of Business Owners: 2002.
Retrieved from https://www.census.gov/prod/ec02/sb0200cscbo.pdf
U.S. Small Business Administration. (2012, September). The Small Business Advocate. The
Small Business Advocate Newsletter, 31(6), 8.
48
van Gelderen, M., Thurik, R., & Bosma, N. (2005). Success and Risk Factors in the Pre-Startup
Phase. Small Business Economics, 24(4), 365–380. Retrieved from JSTOR.
Zarya, V. (2018, January 31). Female Founders Got 2% of Venture Capital Dollars in 2017.
Retrieved October 16, 2018, from Fortune website:
http://fortune.com/2018/01/31/female-founders-venture-capital-2017/
Zolin, R., Stuetzer, M., & Watson, J. (2013). Challenging the female underperformance
hypothesis. International Journal of Gender and Entrepreneurship, 5(2), 116–129.
https://doi.org/10.1108/17566261311328819
Zwan, P., Thurik, R., Verheul, I., & Hessels, J. (2016). Factors Influencing the Entrepreneurial
Engagement of Opportunity and Necessity Entrepreneurs. Eurasian Business Review,
6(3), 273–295. https://doi.org/10.1007/s40821-016-0065-1
