Signalling In Equity-Crowdfunding: An Exploratory Study ...€¦ · Crowdfunding allows early-stage...
Transcript of Signalling In Equity-Crowdfunding: An Exploratory Study ...€¦ · Crowdfunding allows early-stage...
Signalling In Equity-Crowdfunding:
An Exploratory Study Into Herding Behaviour
Amongst First-Time Retail Investors
By
Luke Barratt
610003924
026073
Supervised by
Dr. Boyi Li
Submitted to the University of Exeter Business School
as part of the BUS3001 Dissertation Module
Word Count: 9972
29th of April 2016
1
Abstract
Title: Signalling In Equity-Crowdfunding: An Exploratory Study Into
Herding Behaviour Amongst First-Time Retail Investors
Author: Luke Barratt
Supervisor: Dr. Boyi Li
Keywords: Signalling, Crowdfunding, Equity-crowdfunding, Herding, First-
time Investors, Crowdfunding Investment Criteria
Purpose: This study aims to provide exploratory findings into the relative
importance first-time retail investors place on social signals within
equity-crowdfunding pitches to discover whether they act
independently in their investment valuation or collectively as part
of a herd.
Methodology: The method adopted for this study has firm basis in previous
literature concerning the decision-making criteria used by venture
capitalists to evaluate early-stage ventures as pioneered by
MacMillan et al., 1985. The mixed-method sequential exploratory
approach uses insights from qualitative semi-structured
interviews to construct a quantitative online questionnaire for data
analysis.
Conclusions: Social signals were found to always be an important component of
the sample’s due diligence process. Investor characteristics such as
gender and the importance placed on both managerial and
financial signals significantly influence the relative importance
placed on social signals. However, social signals were of lowest
importance when compared to other categories, providing no
suggestion of herding behavior, implying that first-time retail
investors act independently in their investment valuation.
2
Acknowledgements
Firstly, I would like to thank my supervisor Dr. Boyi Li for his dedication,
guidance, enthusiasm and support throughout the entire span of the project.
I would also like to thank my friends and family for their attentive listening,
insights and unconditional support. The completion of this project would not
have been possible without their help.
To George Cockburn and David Watson for their open-mindedness and interest
during the early stages of the project.
To Tej Panesar, for his helpful discussions and insights in interpreting the
results.
Finally to all interviewees and survey respondents who were kind enough to
dedicate their free time to this study, my sincere thanks.
3
Table of Contents
1. Introduction ...................................................................................................... 5
2. Literature Review .............................................................................................. 7
2.1 Definition And Origins Of Crowdfunding ........................................................................................ 8
2.2 Types Of Crowdfunding .......................................................................................................................... 9
2.3 Role Of Crowdfunding In Entrepreneurial Finance ................................................................. 10
2.4 Overview Of The Equity-Crowdfunding Industry ..................................................................... 11
2.5 Information Asymmetry And Early-Stage Venture Valuation ............................................. 13
2.6 Signalling In Equity-Crowdfunding And The Role Of Online Platforms .......................... 14
2.6.1 Signalling Problems In Equity-Crowdfunding ........................................................................... 14
2.6.2 Information Asymmetry Remedies In Crowdfunding ............................................................ 15
2.7 Social Signalling And Herding ........................................................................................................... 17
3. Method ........................................................................................................... 18
3.1 Research Problem Definition ............................................................................................................ 18
3.2 Research Design ...................................................................................................................................... 19
3.3 Data Collection ........................................................................................................................................ 21
3.3.1 Preliminary Interview .......................................................................................................................... 21
3.3.2 Qualitative Research ............................................................................................................................. 21
3.3.3 Quantitative Research .......................................................................................................................... 23
3.4 Data Analysis ............................................................................................................................................ 25
3.4.1 Qualitative Data Analysis.................................................................................................................... 25
3.4.2 Quantitative Data Analysis ................................................................................................................ 25
4. Findings and Discussion ................................................................................... 26
4.1 Qualitative Research ............................................................................................................................. 26
4.2 Quantitative Research .......................................................................................................................... 31
4.2.1 Data Cleaning ........................................................................................................................................... 31
4.2.2 Data Checks ............................................................................................................................................... 32
4.2.3 Descriptive Statistics ............................................................................................................................. 32
4.2.4 Inferential Statistics .............................................................................................................................. 36
4.2.5 Variable Correlations ............................................................................................................................ 36
4.3 Summary .................................................................................................................................................... 42
5. Limitations ...................................................................................................... 45
4
6. Research Implications ..................................................................................... 45
5.1 Legislation ..................................................................................................................................................... 45
5.2 Practitioners ................................................................................................................................................. 46
5.3 Future Research .......................................................................................................................................... 46
7. Conclusion ...................................................................................................... 47
8. References ...................................................................................................... 48
9. Appendices ..................................................................................................... 56
Appendix 1 – Types of Crowdfunding ................................................................................................... 56
Appendix 2 – Mass Solution ...................................................................................................................... 57
Appendix 3 – JOBS Act ................................................................................................................................. 58
Appendix 4 – Sources of Entrepreneurial Finance .......................................................................... 58
Appendix 5 - Venture Capital and Business Angel Investment Criteria .................................. 58
Appendix 6 – Crowdfunding Investment Criteria ............................................................................ 60
Appendix 7 - Interview Process ............................................................................................................... 62
Appendix 8: List of Elicited Quality Signals From Interviews ..................................................... 64
Appendix 9 - Example Interview Transcript ...................................................................................... 65
Appendix 10 - Open and Axial Codes ..................................................................................................... 68
Appendix 11 - Online Questionnaire ..................................................................................................... 70
Appendix 12 - Data Cleaning ..................................................................................................................... 79
Appendix 13 – Reliability Checks ............................................................................................................ 81
Cronbach Alpha .................................................................................................................................................. 81
Shapiro-Wilk Test .............................................................................................................................................. 81
Appendix 14 – Removed Quality Signals ............................................................................................. 85
Appendix 15 - MacMillan et al.’s (1985) Result Comparison ...................................................... 85
Appendix 16 – Top 20 Highest Ranked Variables ............................................................................ 86
Appendix 17 – Independent-Samples T-Test and Age Groups ................................................... 87
Appendix 18 – Ethics Form........................................................................................................................ 87
5
1. Introduction
Equity-crowdfunding has revolutionised financing for small and medium sized
enterprises (SMEs) by providing an accessible means for all members of the
public to invest in projects of interest. It is currently the second fastest growing
sector of alternative finance (Zhang et al., 2016), and has unlocked
approximately £53.5bn of gross UK household savings worth of potential early-
venture investment (ONS, 2016). Yet, little is known about the signals which
trigger investment, especially amongst first-time retail investors. The extent to
which these investors act independently to interpret quality signals and harness
‘The Wisdom of Crowds’ could determine the success of equity-crowdfunding in
effectively selecting new ventures.
Crowdfunding allows early-stage ventures to raise capital by collecting a series
of small contributions online. “Entrepreneurs make an open call for funding on a
crowdfunding platform, and investors make their decisions based on the
information provided therein” (Ahlers et al., 2015). Equity-crowdfunding
extends this concept and allows investors to receive financial returns in
exchange for providing capital. This provides retail investors or “retail clients
who are neither sophisticated investors nor high net worth individuals... of
ordinary means and experience who make up the vast majority of the retail
market in the UK” (FCA, 2013), the possibility to financially support projects.
However, since crowdfunding is a new phenomenon academic research remains
limited. Whilst prior research concentrates on crowdfunders’ motivations,
further research is necessary to better understand their decision-making process
and specifically the signals which trigger investment (Scheder & Arboll, 2014,
Ahlers et al., 2015), particularly in the equity-crowdfunding context (Burtch et
al., 2013; Mollick, 2014; Kuppuswamy & Bayus, 2015). Knowledge into signalling
is particularly relevant amongst first-time retail investors since they are the
largest investor group by number and, partly due to their lack of expertise, could
sustain substantial losses from investing in high risk early-stage ventures. This is
especially concerning given that equity-platforms do not conform to strict
6
governance and reporting requirements common in security marketplaces
(Schwienbacher & Larralde, 2010; Agrawal et al., 2014). Although legislation
places restrictions on the amount retail investors can invest (FCA, 2015), legal
requirements globally are becoming more liberal.
Consequently, this paper focuses on the quality signals which influence retail
investors’ decision-making when evaluating new venture proposals. The
project’s pitch on a crowdfunding platform is instrumental to investors in
mitigating the information asymmetry associated with assessing the project’s
underlying quality and likelihood of success. These information asymmetries are
enhanced amongst equity-crowdfunding since entrepreneurs must create both a
successful project and generate equity (Agrawal et al., 2014). Furthermore, due
to the unprecedented information provided by equity-crowdfunding platforms,
understanding quality signals could identify common attributes amongst
successful businesses in their early stages.
Various types of quality signals have been identified by investors when
evaluating new ventures, including; managerial (Tyebjee & Bruno, 1984),
intellectual (Agrawal et al., 2014) and financial (Ahlers et al., 2015). However, in
equity-crowdfunding, it is possible to observe social signals in the form of
numerous other investors’ opinions and investments. “Because of such rich
information signals, crowd-funding markets are rife with the potential for social
influence...which can have a very large effect on consumer decisions and product
success” (Burtch, 2011).
However, the significance of this effect is divided amongst crowdfunding
literature. Mollick (2013) suggests that social signals complement investors’
decision-making process, however, they behave as ‘independent amateurs’.
Schwienbacher & Larralde (2010) support this view as investors make decisions
as part of a ‘collective intelligence’ to harness what Surowiecki (2004) coined
‘The Wisdom of Crowds’. Surowiecki (2004) explains that diverse groups, over-
time, make superior decisions than individuals and, “paradoxically, the best way
for a group to be smart is for each person in it to think and act as independently
7
as possible.” Following this view, investors interpret social signals as
complementary in their due diligence process to make high-quality investments.
In contrast, Burtch (2011), Zhang & Lui (2012) and Herzenstein et al. (2011)
maintain that due to the prevalence of social signals investors herd together.
Additionally, Burtch (2011) suggests that this herding effect increases
proportionally to the number of marketplace members and causes investors to
behave irrationally and make poorer decisions as a result. If confirmed, this
could have severe implications given the growth of crowdfunding platforms
worldwide.
This disparity and its potential implications highlight the importance of further
research. Consequently, this study will provide exploratory findings into the
relative importance first-time retail investors place on social signals in their
decision-making process to suggest whether they act independently or as part of
a herd. In order to achieve this, first the relevant literature surrounding
crowdfunding, information asymmetry, signalling in early-stage venture
valuation and herding will be explored. The adopted methodology will then be
justified and described before a presentation of the results. Finally, the study’s
limitations will be addressed before explaining the potential implications for
practitioners and avenues for future research.
2. Literature Review
This literary review provides context to this study by first describing the origins
and types of crowdfunding and their role in entrepreneurial finance before
providing an overview of signalling as a remedy to information asymmetry
problems in early-stage venture valuation.
8
2.1 Definition And Origins Of Crowdfunding
Collectively raising funds from numerous investors is a longstanding concept,
however, with the Internet the process has become more efficient in the form of
crowdfunding. Digital-technologies have enhanced the speed and ability for
society to collaborate and share information (Lehner, 2013). This increased
connectivity gives people access to a wider network of diverse skills,
backgrounds and ideas (Benkler, 2006), which when organised can provide a
valuable and innovative source of labour through a process known as
crowdsourcing (Howe, 2006).
Crowdfunding evolved as an extension to crowdsourcing, where the dispersed
crowd provides capital as opposed to labour (Gerber et al., 2012). Specifically,
“crowdfunding involves an open call, mostly through the Internet, for the
provision of financial resources either in form of donation or in exchange for the
future product or some form of reward to support initiatives for specific
purposes.” (Belleflamme et. al, 2014). Equity-crowdfunding has developed this
concept further “whereby an entrepreneur sells a specified amount of equity or
bond-like shares in a company to a group of (small) investors through an open
call for funding on Internet-based platforms” (Ahlers et al., 2015). Whilst some
academics distinguish between crowdfunding and crowdinvesting, the terms will
be used interchangeably within this paper since crowdinvesting has yet to gain
traction amongst the relevant literature (Appendix 1).
Crowdfunding provides a favourable source of funding for SMEs due to a
reduction in search costs, communication costs and risk as small payments
become feasible online (Agrawal et al., 2014). This, coupled with the recent
global financial crisis and consequent seizure of the predominant, traditional
sources of early-venture funding namely; Venture Capitalists (VCs), Business
Angels (BAs) and bank loans, explains crowdfunding’s rapid growth. Since first
occurring in 1997 (Masters, 2013), crowdfunding is currently the fastest
growing alternative finance area with platforms raising an estimated $16.2bn
globally in 2014 (Appendix 2). Furthermore, the continued global harmonisation
9
of legal requirements allowing retail investors to participate in equity-
crowdfunding, with the recent approval of the JOBS Act Title III in the USA
(Appendix 3), crowdfunding is likely to exceed VC investment in 2016 (Mass
Solution, 2015).
2.2 Types Of Crowdfunding
Given crowdfunding’s broad definition, numerous types have emerged with
different motivations for both investors and entrepreneurs (Gerber et al., 2013).
These can be segmented into donation-based crowdfunding (rewards-based and
charity-based) and investment-based crowdfunding (debt-based, equity-based
and royalty-based) (Appendix 1). This paper focuses on equity-crowdfunding to
extend existing theoretical models concerning the assessment of early-stage
ventures and since the increased information asymmetries, relative to other
crowdfunding forms (Figure 1), provide an ideal environment for evaluating
herding behaviour (Burtch, 2011).
Adapted from: (Hemer et al., 2011)
Figure 1: The Major Forms Of Capital Provision Ranked By Process Complexity
Investments, Equity
Lending
Pre-selling
Sponsoring Donations
Pro
cess
Co
mp
lex
ity
10
2.3 Role Of Crowdfunding In Entrepreneurial Finance The sources of finance available to entrepreneurs and SMEs depend principally
upon the associated risk, which decreases as the business develops (Figure 2).
Schwienbacher & Larralde (2010) highlight two principal sources of finance;
debt and equity, each with numerous subgroups (Appendix 4). VCs and BAs are
generally the preferred funding source since their experience, network and
operational support (Baum & Silverman, 2004) usually provides superior growth
when compared with firms funded by other methods (Keuschnigg, 2004).
However, since these profit-seeking investors are risk-averse they tend to invest
in established businesses. This is predominantly due to the extensive, location-
specific and costly due diligence process in evaluating potential investments.
Furthermore, since the financial crisis “traditional sources of risk capital have
increasingly been moving their investment activity upstream” (Collins &
Pierrakis, 2012), causing businesses to face a scarcity of available capital known
Figure 2: Sources Of Financing In The Entrepreneurial Lifecycle
Taken From: (Scheder & Arboll, 2014)
11
as the ‘ Valley of Death’. This can prevent potentially successful ventures from
progressing, costing the economy jobs and innovation (Bradford, 2012).
To address this funding gap, early-stage projects rely on contributions from
Friends, Family or Founders (FFF) in a process known as ‘bootstrapping’ (Mitra,
2012). However, crowdfunding has become an alternative source of early-stage
capital, due to its ability to allocate the associated due diligence costs and risk
across a larger number of investors (Belleflamme et al, 2010). Agrawal et al.
(2014) explain when entrepreneurs’ potential to use home-equity loans as a
source of financing increases, the number of entrepreneurs who turn to
crowdfunding decreases. Crowdfunding platforms also partially overcome
geographical limitations of traditional finance (Agrawal et al., 2011), although
funding is concentrated in particular areas (Agrawal et al., 2013; Mollick, 2014).
Indeed the ability of crowdfunding to address this previously underserved niche,
explains its rapid growth and the recent relaxations in securities law to allow
retail investors to crowdfund (Parsont, 2013).
Equity-crowdfunding also has potential societal benefits as it enables a more
efficient allocation of financial resources since new ventures are no longer
exclusively chosen by professional investors but by the wider society itself.
Furthermore, crowdfunding has dramatically increased potential funding
available to SMEs through allowing retail investors to invest their savings. SMEs,
although risky, generate a “disproportionately large share of all new net jobs”
(Henrekson et al., 2010:240) and increase productivity in the economy, through
increased innovation, competition and diversity (Carree & Thurik, 2003) in a
process known as ‘creative destruction’ (Brown et al.,2014).
2.4 Overview Of The Equity-Crowdfunding Industry The global equity-crowdfunding industry is lucrative and growing rapidly,
raising $1.1bn with a growth rate of 182% in 2014 and is expected to overtake
VC funding in 2016 (Appendix 2). It currently accounts for 15.5% of the UK seed
and venture stage investment market, although its share is growing quickly
(Zhang et al., 2016).
12
There are two principle platform business model types; investor-led and
company-led. On investor-led platforms, such as SyndicateRoom, a lead investor
negotiates investors’ payment terms with the venture. Company-led platforms,
such as Crowdcube, allow the venture to set their own payment terms (Smith,
2015). Platforms generally operate an all-or-nothing system, whereby a project
only receives investment once its target goal has been reached, although projects
can overfund at the venture’s discretion.
Platforms receive revenue through three principal streams; interest on capital
pledged until a project reaches its goal, a transaction and payment processing fee
(usually 5 % of total funding raised), and an administration fee of approximately
£1750 per campaign (Belleflamme et al., 2015). Crowdcube is used for this study
as it is the largest equity-crowdfunding platform, in terms of both registered
investors (273,244) and accumulated capital (£155 million) as shown in Figure 3
(Crowdcube, 2016).
Figure 3: Global Equity-Crowdfunding Platforms By Capital Raised
Taken from: (Crowdsurfer, 2016)
13
2.5 Information Asymmetry And Early-Stage Venture Valuation
Information asymmetry refers to transactions where one party has superior
information than the other. This is particularly relevant to early-stage venture
valuation as the complexities associated with evaluating the potential for future
business success depict that entrepreneurs have significantly better information
about a venture’s true value than investors (Agrawal et al., 2011). This imbalance
can cause adverse selection, whereby a party only chooses investments which
benefit them the most, at the other party’s expense. Akerlof (1970) famously
described the phenomenon as “The Market For Lemons”, whereby low market
prices drive away sellers with high-quality goods leaving only low-quality
products or ‘lemons’ behind.
Consequently, since the quality of a venture cannot be observed directly,
investors assess its value based upon observable signals which are indicative of
high-quality. Effective signalling can mitigate the market failure of adverse
selection as it allows sellers, or entrepreneurs in this context, to communicate
their quality and derive a fair capital price. However, for successful signalling,
the costs of emitting the signals must not outweigh their benefits and the
transferred information must be interpreted of value by the receiver or, in this
case the investor, so that misleading signals are not rewarded (Connelly et al.,
2011).
Intuitively, not all the information on a venture’s quality will ultimately be an
effective signal. However, VCs and BAs have recognised numerous effective
quality signals in their due diligence process (Appendix 5) and various attempts
have been made to categorise these signals. Tyebjee & Bruno (1984) suggested
expected return (market attractiveness, product differentiation) and perceived
risk (managerial capabilities, environmental threat resistance, cash-out
potential). However, Baum & Silverman (2004) offer more objective
classifications of human capital, social (alliance) capital and intellectual capital.
Although individual signals’ importance are contested, managerial capabilities
are consistently perceived as most important, ahead of financial considerations
14
and product or market characteristics (Tyebjee & Bruno, 1984; Macmillian et al,
1985; Unger et al., 2011). Signals can successfully resolve information
asymmetry problems, with Zacharakis & Meyer (2000) even suggesting that an
actuarial approach relying solely on effective signals can outperform VCs in
choosing profitable ventures.
2.6 Signalling In Equity-Crowdfunding And The Role Of Online Platforms The following section will first describe the information asymmetry problems in
equity-crowdfunding, before describing the potential remedies and social
signalling in particular.
2.6.1 Signalling Problems In Equity-Crowdfunding Information asymmetries are pronounced in equity-crowdfunding since
information is transferred digitally, making interpretation and monitoring
difficult. Consequently the adverse selection risk is increased since investors
must assess the entrepreneur’s ability to deliver both the project and generate
equity (Mollick 2014). Equity-crowdfunding also provides multiple disincentives
to entrepreneurs’ including; the opportunity cost of advice from VCs and BAs,
investor management and competitive risks of information disclosure (Argawal
et al., 2013). Additionally, due to infrequent interactions and consequent
inability of investors to influence entrepreneurs, moral hazard1 in the form of
fraud or demotivation becomes a concern (Mollick, 2014).
Online platforms, which control the interaction between investors and
entrepreneurs, play a vital role in mitigating these market failures by providing a
medium for signalling. However, the platform’s incentives to attract as many,
large, fully-funded projects as possible may not be aligned with those of
investors (Belleflamme et al., 2015), with recent reports suggesting platforms
manipulate pitch information to trigger investment (Hurley, 2016).
1 A situation where one party takes excessive risk because another party bears the costs of the risk
15
2.6.2 Information Asymmetry Remedies In Crowdfunding There are principally three potential remedies to information asymmetry in
crowdfunding; rules and regulation, reputation signalling and crowd due
diligence.
Rules and regulation limit risk exposure by improving the quality of information
transferred through increasing disclosure from platforms and restricting
investors’ leverage (Mollick, 2014). Currently, platforms must notify investors of
the associated risks (FCA, 2015), however, they have limited accounting and
reporting requirements in comparison to other security marketplaces
(Scheinbacher & Larralde, 2010) and legal requirements globally are becoming
more liberal (Pricco, 2015).
Consequently, reputation signalling through the project pitch displayed on the
platform has increased relevance in enabling investors to evaluate potential
investments (Connelly et al., 2011). Crowdfunding investors respond to quality
signals in a similar way to VCs and BAs (Mollick, 2013; Gunther et al., 2015)
demonstrating that they act rationally in their due diligence process, “regardless
of their expectations for financial return” (Mollick, 2014). Equity-crowdfunding
investors also use similar signals in their due diligence process (Appendix 6).
These have most recently been categorised into ‘fact-based signals’ of venture
quality (Human, social (alliance), intellectual) and ‘performance based signals’
which imply the level of uncertainty (equity share and financial projections)
(Ahlers et al.,2015). However, the relative weighting of both individual signals
and signal categories is ambiguous and under-researched.
Previous research has primarily focussed on the unique, differentiating factor of
crowdfunded markets; the aspect of crowd due diligence through social signals.
As Burtch (2011) explains, “the presence of rich, publicly observable information
on prior others’ investment decisions... provides potential investors not only
with an indication of whether others’ invested in a project, but also the timing of
that investment.” Further examples include; online discussion and debate
(Gerber et al., 2012), number of investors (Ahlers et al., 2015), speed of
16
investment (Crosetto & Regner 2014), experience of investors (Zhang & Lui,
2012) and social shares (Burtch et al., 2013).
These indicators reflect mass confidence which validates the project’s concept
and can imply future market success (Schweinbacher & Larralde, 2010).
Investors could interpret these social signals as a complementary component of
their due diligence process to resolve information asymmetry problems and
select high-quality ventures by highlighting issues such as fraud (Mollick, 2014).
For example, early investments may signal entrepreneurial commitment as a
project may be of low-quality if it fails to reach any funding from its founders,
friends and family (Agrawal et al, 2011). This phenomenon, ‘The Wisdom of
Crowds’, highlights the potential of equity-crowdfunding to outperform VCs and
BAs in evaluating new-venture quality as diverse groups make superior
decisions than individuals since they incorporate many different perspectives
and are immune from individual biases (Surowiecki, 2004).
However, for collective intelligence to be successful all members of the crowd
much act independently in their investment decision. Otherwise, investors can
be influenced by social signals to follow prior investors’ decisions to make
irrational and potentially harmful choices in a phenomenon known as herding
(Burtch, 2011).
Although studies have shown that social signals are an important factor in
equity-crowdfunding investors’ decision-making processes (Hornuf &
Schwienbacher, 2014; Kuppuswamy & Bayus, 2015) their relative importance
remains disputed. Mollick (2013) and Schwienbacher & Larralde (2010) suggest
that they are interpreted as complementary to other quality signals whereas
Burtch (2011) implies that they outweigh other signals.
17
2.7 Social Signalling And Herding
In addition to solving information asymmetries, social signals have a distinct
function in crowdfunding given the all-or-nothing platform model dictates that
investors must act collectively to reach the funding target for a project to be
successful. Consequently, quality signals are magnified through a Matthew Effect
(Merton, 1968) since high-quality projects attract investors who further promote
the project to external media or through social sharing (Burtch et al., 2013). This
multiplier effect depicts that identifiable signals of project quality are predictors
of project success (Mollick, 2014).
Following this, since prior investors’ investments provide a credible quality
signal to potential investors, herding can occur as investors may simply support
projects that others have already supported. This is because as the number of
previous investors increase, potential investors become less responsive to their
own information in the belief that others have superior information (Banerjee,
1992). Consequently, first-time retail investors are more likely to herd as their
inexperience makes them more inclined to rely on the information of others
(Kim & Viswanathan, 2016). Zhang & Liu (2012) explain although herding is
usually associated as being irrational with negative consequences (Burtch, 2011)
rational herding can be beneficial as high-quality projects receive
disproportional media attention and investment which can entice others to
invest.
Investors interpret level of funding differently depending on the type of
crowdfunding (Kuppuswamy & Bayus, 2015). Although Zhang & Liu (2012),
Herzenstein et al. (2011) and Burtch (2011) suggest investors herd in equity and
lending-based crowdfunding, “there is very little empirical research to
definitively support any position.” (Kuppuswamy & Bayus, 2015). Indeed, some
studies have found the contrary, recognizing a Bystander Effect whereby
investors contribute smaller amounts less frequently as a project approaches its
target funding amount believing that others will contribute instead
(Kuppuswamy & Bayus 2015; Burtch et al., 2013).
18
To summarise, social signals play an important role in investors’ decision-
making processes and can influence the likelihood of funding success. However,
the weight individual investors attach to social signals is not well understood. If
social signals are perceived as having superior importance to other signalling
categories, herding could occur. However, if social signals are perceived as
having less relative importance, investors use them as complementary indicators
of a venture’s quality in conducting their own due diligence behaving as
‘independent amateurs’ to harness ‘The Wisdom of Crowds’. Consequently, this
study aims to provide insights into the relative importance first-time, retail
investors attach to social signals to determine if they act independently or as
part of a herd.
3. Method This section will first define the research problem, before justifying the research
design and detailing the data collection and analysis processes.
3.1 Research Problem Definition The herding phenomenon will be explored by evaluating the importance of social
signals in first-time retail investors’ due diligence process when compared
against other signal categories. However, given that importance is a relative
measure it is difficult to assess through qualitative research alone. Similarly,
weighting of signals through a quantitative approach would not have much value
as it is unknown whether the signals are credible or interpreted positively or
negatively. These inherent complexities of signalling and herding depict that
research is naturally suited to a mixed-method approach.
Consequently, this study uses qualitative interviews to compile potential signals
used by first-time retail investors to evaluate venture quality from observable
information on the project pitch, presented through the Crowdcube platform. A
quantitative online questionnaire was then distributed to determine the
importance the sample attached to each individual quality signal. Finally,
insights from the interviews and previous literature were used to categorise the
19
signals for data analysis and comparison. The resulting categories used for the
research were; product, managerial, financial and social.
These categories were chosen to provide research continuity and derived
principally from Baum & Silverman’s (2004) VC and BA investment classification
of human, intellectual and social (alliance) capital, which have been adopted in
equity-crowdfunding studies (Ahlers et al., 2015). However, to classify the
findings more in line with industry rating agencies (CrowdRating, 2016),
intellectual capital was broadened to include both product and market
characteristics as from Macmillan et al.’s (1985) study and a financial category
was created to deal with financial characteristics (MacMillan et al., 1985) and the
level of uncertainty (Ahlers et al., 2015).
Broad categories where chosen to simplify comparisions and signals were
allocated to each category based on insights from previous theory, at the
researcher’s discretion. This highlights a limitation of early-stage venture
signalling research because although signal categories are fairly uniform the
individual signals within those categories are not. For a list classified signals for
the quantitative research refer to Figure 5 and for removed or unclassified
signals refer to Appendix 14.
3.2 Research Design Qualitative research is the most appropriate way to explore new fields of
research as basis for further inquiry (Edmondson & McManus, 2007). An
exploratory sequential mixed-method approach is particularly useful to identify
important variables to study quantitatively when the variables are unknown and
when exploring a phenomenon to measure its prevalence (Creswell, 2013).
Consequently semi-structured interviews were used to elicit quality signals
before conducting statistical testing, as previously adopted and recognised in the
VC and BA context (MacMillan et al., 1985; Tyebjee & Bruno 1984). Adopting the
same approach allows for continuity throughout signalling and early-venture
valuation research. Furthermore, asking interviewees to discuss their
20
interpretations of the project’s quality rather than their motivations to invest,
increased the validity of findings by avoiding the issues associated with vignettes
or hypothetical scenarios (Barter & Renold, 1999).
Furthermore, since the less we know about a field the more open data collection
needs to be (Edmondson & McManus, 2007), especially given the dangers of
researcher or existing theory bias (Bryman & Bell, 2015) a grounded theory
approach was adopted when uncovering credible quality signals. This was
especially relevant, when considering the subjective interpretation of signals
depending on the individual (Connelly et al., 2011) and crowdfunding context
(Kuppuswamy & Bayus, 2015) and since no widely recognised theory exists as to
how first-time retail investors interpret quality signals in equity-crowdfunding.
An iterative interviewing process was used so each additional interview could
build upon previous research to create a comprehensive list for quantitative
analysis and provide insights into the general consensus of the sample’s
interpretation of particular signals (Daymon & Holloway, 2010).
Incorporating an online questionnaire developed the conceptual framework of
the decision-making process by providing a more rigid basis for quantifying the
importance of quality signals so they could be compared against one another and
in other crowdfunding contexts. It complements the interviews as interviewees
tend to have difficulties accurately expressing themselves in conversation and
quantitative research alone fails to determine the logical sequence of events in
decision-making processes (Creswell, 2013). Consequently, the more structured
approach in the questionnaire by asking respondents to inspect quality signals
rather than evoke them from pitch information enabled them to consider their
significance more thoroughly, thereby improving validity.
The methodology presented in this study is applicable to situations where
researchers need to develop, test and compare unknown variables to determine
their significance. It has clear relevance in assessing early-stage venture quality
signals (Macmillan et al., 1985; Tyebjee & Bruno 1984), which can be explored
further and compared in other crowdfunding contexts and investor groups, but
also has considerable potential applications for other areas. Its application in
21
marketing could be of particular significance given the complexities of assessing
the value customers place on different aspects of a product or service.
3.3 Data Collection Data collection was designed to give a comprehensive overview of signals used
in first-time retail investors’ equity-crowdfunding venture assessment, through
incorporating multiple perspectives and methods. Data collection was conducted
in a three-tiered process; a preliminary interview, qualitative semi-structured
interviews and quantitative online questionnaire.
3.3.1 Preliminary Interview A preliminary semi-structured interview was conducted with George Cockburn
and David Watson from the Annual Fund at Exeter University, who run a
donation-based crowdfunding platform. The principal aim was to gain a
practitioner’s perspective of crowdfunding platforms’ due-diligence process and
advice given to entrepreneurs as to the most important quality signals which
trigger investment.
Although specific to donation-based crowdfunding, useful insights were obtained
into; investor behaviour, the importance of the pitch video in generating project
interest and the influence of social media on project success. These perspectives
were incorporated throughout the remaining data collection process.
3.3.2 Qualitative Research Qualitative research data was obtained through 16 semi-structured interviews,
over a period of two-weeks, with the aim of gathering a comprehensive list of
quality signals used in first-time retail investors’ due diligence process and, most
importantly, insights into how those signals are interpreted. The grounded
approach depicted that signals were only documented once they were
mentioned by the sample, eliminating potential biases from previous research
(Edmondson & McManus, 2007).
22
The sample was constructed through convenience and snowball sampling, given
the limited resources available to the researcher. Respondents were screened to
ensure they had never previously invested in a crowdfunding campaign, and
were defined as retail investors by the Financial Conduct Authority (2013).
Although the sample had a relative gender bias (10 female and 6 male) and was
homogeneous in terms of age group (19-26) it was appropriate given the
exploratory focus of the study, and due to data saturation fewer original signals
were collected with each additional interview (Guest et al., 2006).
The interview process was intended to replicate a genuine investment and
consisted of three stages. Interviews were conducted at the participants’ home,
where they would most likely make online investments. Before proceeding, the
interviewees were offered confidentiality, regarding their responses and
participation in the study, and asked if they were comfortable with the equity-
crowdfunding process. Where applicable, their queries were discussed with the
interviewer.
Participants were then given a laptop and directed to the crowdcube.com
homepage. They were asked to browse current investment opportunities until
they selected a project of interest. The interviewer observed and noted their
browsing behaviour and then asked the interviewees to analyse all the
components of the investment opportunity’s pitch. Examples of browsing
behaviour included the use of filters to refine project options, scrolling speed and
comments on aspects of project pitches. The purpose of observing participants’
browsing behaviour was to see whether popularity listings influenced decision-
making as found by Burtch (2011) and provide a more comprehensive list of
signals since they can be interpreted subconsciously (Connelly et al., 2011).
Interviewees were asked to recall their selection-process and identify and locate
which aspects of the investment opportunity indicated the project’s quality.
Interview responses were taped to capture interpretations as effectively as
possible and considered appropriate given the non-confidential material under
23
consideration and volume of signals would have been difficult to document by
note-taking alone (Walsham, 1995).
It was essential to ask participants to recall information directly after assessing
the pitch as participants’ ability to accurately recall their decision-making
process diminishes over time (Ramser, 1993). Points of discussion evolved with
each additional interview with a final list of prompt questions and the
interviewing process found in Appendix 7.
3.3.3 Quantitative Research The quantitative research and online questionnaire developed the
understanding of the elicited quality signals and created a more comprehensive
framework of respondents’ decision-making process. A four-point Likert-scale
was incorporated from Macmillan et al.’s (1985) study to form the basis for
statistical analysis.
The online questionnaire was also designed to replicate a genuine investment.
The principals of equity-crowdfunding were explained to participants and they
were asked to give consent for their anonymous data (age and gender) to be
used for research purposes. Participants were also asked whether they had
previously invested in a crowdfunding campaign and whether they were retail-
investors as defined by the Financial Conduct Authority (2013). Previous
crowdfunding investors, high net-worth individuals and sophisticated investors
were, therefore, excluded from the responses.
Participants were then presented with a complete copy of a randomly chosen
project selected from the Crowdcube website. Aspects of the project pitch, which
were found to be confusing to interview respondents, were annotated to help
clarify their meaning. It is important to note that although the pitch information
was included in the questionnaire it was not presented in the exact same format
as on the Crowdcube platform which may alter interpretations. Despite this
limitation, it ensured dynamic elements of the pitch such as the level of funding
24
were held constant so respondents were referring to the same information when
evaluating the project.
Participants were then asked to rate all 73 quality signals according to their level
of importance using a four-point Likert-scale (Figure 4). Quality signals were
incorporated into the model if one or more of the interviewees mentioned it as
being significant, with a final list in Appendix 8. Where applicable, the phrasing
of questionnaire questions was kept constant with previous research into VCs’
and BAs’ decision-making criteria in order to provide reliable comparisons.
Sampling was also a combination of convenience and snowballing, as the
questionnaire was distributed through social media and participants were asked
to encourage others to take the survey in order to increase responses.
Interviewees were asked to complete the survey in order to provide continuity
with the results, although given the anonymity of questionnaire responses, it is
unknown if they did. Online distribution was considered appropriate, as
crowdfunding is an online process by nature, and allowed for a diverse group of
survey respondents. Screenshots of the questionnaire are in Appendix 11.
Figure 4: Likert-Scale Criteria For The Online Questionnaire
Taken from: (Macmillan et al., 1985)
25
3.4 Data Analysis This section details both the qualitative and quantitative data analysis processes.
3.4.1 Qualitative Data Analysis Due to the volume of signals mentioned in the interviews an open-coding
technique was adopted on the transcribed interviews using NVivo qualitative
data software. An example transcript can be found in Appendix 9. This software
was chosen on the basis that the software’s features were appropriate for the
study’s aims. Once all signals mentioned had been identified, axial coding was
used in conjunction with existing VC, BA and Crowdfunding theory to categorise
the signals as suggested by Orlikowski (1993) and are included in Appendix 10.
3.4.2 Quantitative Data Analysis Once the signals were collected and categorised, their relative importance was
assessed by assigning values 1-4 to the four-point Likert-scale (Figure 4).
Statistical analysis was conducted through IBM’s SPSS software to display the
mean, mode and standard deviation of each variable. The SPSS software was
chosen since its simple interface and available support material allowed the
inexperienced researcher to conduct complex statistical data manipulations.
These analyses developed the conceptual framework into the sample’s decision-
making process by quantifying the relevant importance of each signal and their
relationship in regards to other signals.
To analyse the relationships between signal categories, the original 73 signals
were refined to 58 signals. The quantitative research provided a consistency
check of the elicited signals as, following Macmillan et al.’s (1985) approach, all
signals with a mean less than 2 were removed from the study. This was
appropriate as signals below this mark were likely to distort findings, as they
were considered generally irrelevant by the sample. Additionally, signals that
could not be appropriately allocated to one of the categories in the model;
product, management, financial or social were also removed given that their
26
inclusion would increase the complexity without necessarily improving insights
into the focus of this study; the relationships between signal categories.
The resulting descriptive data was then checked for reliability and normality to
determine its appropriateness for parametric testing. A Paired-Samples T-Test
was used to uncover category relationships, and an Independent-Samples T-Test
to uncover relationships with regard to age or gender. Finally a standard
multiple regression was constructed to measure the strength of category
relationships when controlled for gender.
4. Findings and Discussion The following section discusses the significance of the study’s findings into the
relative importance first-time retail investors place on social signals when
evaluating the quality of equity-crowdfunding pitches. The results of the semi-
structured interviews will be discussed first before a presentation of the
quantitative data.
4.1 Qualitative Research In addition to provide a comprehensive list of potential quality signals for
statistical analysis, qualitative research was designed to deepen the
understanding of the sample’s interpretation of individual signals. Consequently,
the following themes convey interpretations and avenues for future research,
which might be difficult to obtain from quantitative research alone. Quotes were
selected based on their ability to succinctly reflect the general consensus of the
sample. A list of the most frequently discussed and coded signals of quality is in
Appendix 10. Although this does not reflect the relative importance of each
individual signal it provides insights into the most frequently noticed signals.
27
Theme 1: Browsing behaviour and project choice
Respondents consistently filter projects in terms of their popularity on the
platform and their desired areas of interest. This points to a Matthew Effect as it
appears that popular projects are more likely to get noticed as recognised by
Burtch (2011). The majority of respondents also scrolled through projects
relatively quickly signifying the importance of the project to convey its concept
succinctly to engage interest. The resounding, deciding factor of project choice
was the respondent’s personal interest in the project concept. Although
philanthropic projects were a consistent area of interest, it was not a
requirement as reflected by the type of projects chosen (8 philanthropic projects
out of 16).
Theme 2: Trust of entrepreneur and management team
All respondents mentioned trust as a key area of concern, given that information
asymmetry problems are especially pronounced in the equity-crowdfunding
context due to the issues of fraud as explained by (Mollick, 2014). Although
many respondents recognised the difficulty of conveying trust through the pitch,
the project video was considered the most important means of transmitting
some key signals of trust; the entrepreneur’s personality and preparedness.
“I think you gauge trust from your initial reaction and gut feeling”
- Respondent 7
“Whoever you are investing in should be driven, enthusiastic, knowledgeable and
excited about what they are doing.”
- Respondent 9
“You wouldn’t give your money to someone who didn’t have a clear, viable vision
and plan for how the business will improve.”
-Respondent 13
28
Theme 3: Importance of financial returns
Although many respondents had issues understanding the structure of financial
returns, as can be expected from inexperienced investors, equal financial returns
were phrased essential whilst competitive or superior returns were merely
desirable:
“I think if you are going to invest you need to get positive returns and at the very
least your money back. You can’t just afford to throw away money. Although I
wouldn’t mind if I just got my money back and I was helping a good cause.”
-Respondent 2
“I’d see it more as charity than pure investment. If I were looking for an investment
then there are safer and more profitable ways to go about it.”
-Respondent 5
This further suggests equity-crowdfunding as separate from charitable-
donations and implies equity-crowdfunding investors are less risk-averse than
traditional funding sources cementing the findings of Scheder & Arboll (2014) in
the role of equity-crowdfunding in entrepreneurial finance. This view of equity-
crowdfunding as complementary to VC and BA funding was also supported by
Tej Panesar Head of Credit and Equity Risk at Crowdcube (T. Panesar, personal
communication, April 19, 2016).
Theme 4: Largest investment as an ambiguous signal
The largest investment was often interpreted ambiguously, with some
respondents seeing it as a positive signal which increases the likelihood of
investment, whilst most recognised the potential conflict in interest.
“That would give me a lot of confidence as when you’re investing that kind of
money (£150,000) you’d do a lot of research on the project.”
- Respondent 1
29
“You’d obviously have to question who it was that gave that investment, if it was a
previous investor that was happy with his investment and now wants to contribute
more that would be a different matter to if it was a family friend.”
-Respondent 3
“If there is no connection then I would see that as a very positive sign. However, if
they were very close to the team then I would think that it would negatively impact
my view of the project.”
-Respondent 15
This could be of particular interest to legislative bodies, although according to
Crowdcube providing the largest investment amount is encouraged and projects
must only disclose self-funding if a director contributes more than 10% of the
funding goal (T. Panesar, personal communication, April 19, 2016).
Theme 5: Importance of own due-diligence and platform reputation
Although platforms’ due-diligence was often considered sufficient in verifying
project legitimacy, most respondents suggested that their own due diligence
would be an essential part of their decision-making process.
“I wouldn’t actually check (pitch information), I would trust it on the platform
because I think it’s already been double checked and the amount I’m investing in it
would be small.”
-Respondent 11
“You’d always want to validate everything before investing, although it depends
upon the amount of money invested. I’d definitely want to research them
independently and then perhaps contact them later.”
-Respondent 7
30
Although this supports Mollick’s (2014) view that investors behave like
‘independent amateurs’ it also raises legislative concerns about the relative lack
of restrictions on crowdfunding platforms as discussed by Agrawal et al. (2014).
Theme 6: Homogeneity as a signal of poor-quality
Diversity was another ambiguous signal as it was rarely recognised as a potential
sign of quality, rather homogeneity was considered a signal of poor quality.
“Having a diverse team widens your chance of investors finding a connection with
the team or product.”
-Respondent 14
“As long as there is a good range of experience than that is good enough, I don’t
think it would matter if they were diverse unless it was obviously extremely
homogenous.”
-Respondent 3
These views reflect the recent research of Abrahams (2016), which suggests
board level homogeneity as a potential indicator of poor company performance.
Theme 7: Social signals
Whilst social signals, particularly the level of funding, were considered to
improve trust in the project as found by Ahlers et al. (2015) an overwhelming
majority explained that the project concept and personal opinion were the
overriding triggers for investment.
“Same with most things if there is a lot of positive feedback from other people you
are more likely to trust it and get involved.”
-Respondent 13
31
“If they hadn’t raised much money then I would question why, obviously you would
make considerations for the length of time of the project.”
-Respondent 10
“Well at the end of the day, for it to be successful you need a lot of investors which
share the same opinion. But I think the reality is that no one else’s opinion other
that your own really matters.”
-Respondent 4
As succinctly explained by Respondent 4, these responses provide no inclination
of herding behaviour.
4.2 Quantitative Research In this section data cleaning techniques and data checks will be explained before
presenting the descriptive data and inferential statistics results.
4.2.1 Data Cleaning Given that the questionnaire was of significant length (approximately 15
minutes) and distributed freely through social media it was sensitive to
dishonest responses (Van Selm & Jankowski, 2006). Consequently, a
combination of data cleaning techniques were used to increased the validity of
findings including; screening for straight-lining, Christmas-tree behaviour,
inconsistent answers and speeding (Appendix 12). Of the 179 survey responses,
42 were partially completed, 12 were disqualified according to their investor
classification and 22 were removed for speeding resulting in 103 valid, complete
responses, which were used for analysis. This was an appropriate amount given
the exploratory nature of the findings and in line with previous research;
Macmillan et al. (1985) had 102 responses.
32
4.2.2 Data Checks A Cronbach Alpha test was conducted to test for internal consistency and inter-
relatedness amongst the variables used to measure the importance of each
quality signal (Appendix 13). The score of 0.86 is considered to show good
reliability and sits comfortably above the acceptable benchmark of 0.7 (Travakol
& Dennick, 2011), suggesting that respondents answered the questionnaire
consistently and accurately.
A Shapiro-Wilk test was also used to test whether the sample came from a
normally distributed population. At the p<0.05 significance level, the hypothesis
that the sample comes from a population which has a normal distribution cannot
be rejected, suggesting parametric statistical analysis will provide accurate
results when used in this study. Although, interestingly, the financial category is
close to being classified as a non-normal distribution (Appendix 13).
4.2.3 Descriptive Statistics The descriptive statistics from the valid questionnaire responses can be found on
the following page in Figure 5 and uncategorised or ‘irrelevant’ signals in
Appendix 14. For reference as to values’ significance refer back to Figure 4.
33
Figure 5: Online Questionnaire Descriptive statistics
34
The results appear to be consistent with previous VC and crowdfunding
literature, especially with regards to the significance of the management team.
Given that the same approach was used particular parallels can be drawn with
Macmillian et al.’s (1985) findings (Appendix 15). One notable exception is that
financial signals seem to be of less importance to the sample as can be expected
given their smaller contribution amounts.
Referring back to the focus of this study and the relative weighting placed on
each signal category, the number of variables per category (Figure 6) could be of
significance as it gives an indication of the amount of signals considered
important in the decision-making process. Following this, it appears that many
different social signals are considered by first-time investors; notably more than
both product and financial signals.
However, when considering the average mean importance placed on signal
categories, social signals are the lowest as shown in Figure 7.
Figure 6: Number Of Variables By Category
Figure 7: Mean Signal Importance By Category
35
This depicts that although there is a high quantity of social signals their relative
importance in the decision-making process is low, tending towards ‘desirable’ as
opposed to ‘important’ as is the case with the other categories. Referring back to
Figure 5, social signals also have the lowest variance (0.641) compared to
financial (0.782), management (0.786) and product (0.742), suggesting that the
sample not only finds social signals the least important but are also relatively
consistent in that interpretation.
The importance of the management category becomes clearer when evaluating
the criteria considered most essential (Figure 8). 73.8% of respondents marked
entrepreneur’s trustworthiness as essential, partially signalled by their level of
project involvement (81.6%) and strategy (58.3%), which is consistent with the
theme of management trust as surfaced in the interviews. Furthermore, the
value of the entrepreneur is particularly evident and is consistent with the
Macmillan et al.’s findings (Appendix 15).
Unsurprisingly, as a result the role of the entrepreneur and management
category commonly feature amongst the highest ranked variables (Figure 9).
Figure 8: Top 10 Variables Most Commonly Perceived Essential
36
With regards to social signals only two appeared in the top 20 (Appendix 16)
with only one considered important; ‘investors seem excited by the project’ with
a mean of 3.00. This supports the mean category results in stating that social
signals are of lowest importance relative to other categories, especially given
that of all social signals, when refering back to the Likert-scale definitions, only
one signal would need to be compensated for by other factors for an investment
to take place.
4.2.4 Inferential Statistics Inferential statistics were used to assess and identify patterns amongst the
relationships between signal categories. First a Paired-Samples T-Test was
conducted, followed by an Independent-Samples T-Test and standard multiple
regression.
4.2.5 Variable Correlations Consequently, the mean importance per respondent per category was used to
conduct a Paired-Samples T-Test to better understand the relationships between
categories. The mean score for the social category was M=2.41 (SD = 0.38),
M=2.92 (SD = 0.28) for the management category, M=2.66 (SD = 0.29) for the
product category and 2.54 (SD = 0.41) for the financial category (Figure 10).
Figure 9: Top 10 Highest Ranked Variables
37
When comparing the mean differences between categories using a Paired-
Samples T-Test (Figure 11) it becomes apparent that the social category was
perceived as significantly less important to the financial category, t(102)= -3.35,
p<0.01. It was also significantly less important to the management category,
t(102)= -14.83, p<0.01 and the product category, t(102)= -6.47, p<0.01. In other
words, the null hypothesis that the relationships between categories are caused
by chance or sampling error can be rejected at the p<0.01 significance level,
implying that respondents’ interpretations of social signals are always
outweighed by signals in either of the other categories.
Another key finding is that the category means are positively correlated with one
another; social and financial (0.463), social and management (0.477) and social
and product (0.336), as shown in Figure 12.
Figure 11: Paired-Samples T-Test Results
Figure 12: Paired-Samples T-Test Correlations
Figure 10: Paired-Samples T-Test Statistics
38
The strength of the correlations can be better illustrated, using Scatter Plots in
Figure 13. This depicts that as a respondent’s importance in either the financial,
management or product category rises, a higher importance will be placed on
social signals and can be explained by investors’ varying risk appetite.
Additionally, given that there are no negative correlations, respondents will
always place importance on social signals regardless of their interpretations of
signals in other categories. Social signals are, therefore, always considered in
first-time investors’ decision-making process.
Figure 13: Category Correlation Scatter Plots Using Mean Importance Per Respondent Per Category
39
To further explore data patterns, Independent-Samples T-Tests were used to
segment respondents according to age and gender. Age was not found to reveal
any statistically significant findings (Appendix 17), however, gender held some
interesting characteristics (Figure 14).
These results indicate that females place a higher mean importance on all
signalling categories, implying that they are more risk-averse. Regarding the
Likert-scale definitions, it follows that females are more likely to place signals as
either ‘important’ or ‘essential’ and are less likely to invest as a result. These
findings could partially explain the gender bias (25% female) amongst investors
on the Crowdcube platform (Crowdcube, 2016).
Even more interestingly, when the assumption of homogeneity of variances
amongst social signals was tested using the Levene’s Test (Figure 15) it was
found to be statistically significant F(101)= 0.82, p= 0.37, t(101)= -2.04, p= 0.04.
These findings may be of interest to signalling and herding theorists in particular
as the sample implies that females are more inclined to place importance on the
opinions of others than males.
Figure 14: Independent-Samples T-Test By Gender Statistics
40
Figure 16: Multiple Regression Variable Correlations
Furthermore, the gender difference was also nearly of statistical significance in
the management category F(101)= 0.05, p= 0.82, t(101)= -1.83, p= 0.07. These
results are consistent with research into gender differences as females have a
“tendency to agree more readily with others” (Eagly, 2013:98), partially
explaining why they appear to place more importance on the signal categories
relating to human relationships namely; management and social.
Finally, a standard multiple regression was conducted to further explore the
unique contribution of the financial mean, product mean and management mean
to predict the mean importance of social signals (Figure 16) when controlled for
gender. In the gender dummy variable females were coded as 1, males as 0. Age
was not significantly correlated, and so was removed from the model.
Figure 15: Independent-Samples T-Test By Gender Results
41
The results of the regression (Figure 17) show that the model is a significant
predictor of the mean importance of social signals, F(4,98)= 11.531, p< 0.01.
Furthermore, although the model is a relatively poor fit of the data, R2adj=0.292,
it explains 29.2% of the variance in the mean importance of social signals (Figure
18).
From analysing Figure 19, there are no signs of multicollinearity within the
model, given that all variables have a Tolerance of greater than 0.1 and Variance
Indication Factor (VIF) of less than 10, explaining that each variable in the model
makes a unique and valuable contribution. The management category was found
to be best at predicting the mean importance placed on social signals, ΒM= 0.377,
t= 2.789, p= 0.006, although the financial category was also significant, ΒF=
0.258, t=2.717, p= 0.008.
Figure 19: Multiple Regression Results Coefficientsa
Figure 17: Multiple Regression Results ANOVAa
Figure 18: Multiple Regression Results Model Summaryb
42
The standardised beta, which measures the relative importance of each
independent variable, reflects this indicating that management (BetaM=0.278) is
greater than the financial category (BetaF=0.273). In other words, the mean
importance investors’ place on management and financial categories can indicate
their interpretation of social signals.
The unstandardised beta, explains the expected effect of a one-point increase in
the independent variable on the dependent variable, when all other independent
variables are held constant. Consequently, the management’s ΒM= 0.377 explains
that for every 1.00 increase in the mean importance placed on managerial
signals, the mean importance placed on social signals is expected to rise by
0.377, according to the Likert-scale definitions (Figure 19). Therefore, the model
confirms these correlations found in the Paired-Samples T Test, but in greater
detail by suggesting the extent of the correlation. Interestingly, gender as a
dummy variable was not found to be a significant predictor when all other
independent variables were held constant, ΒG= 0.095 t=1.461, p= 0.147,
suggesting that the effect is smaller than proposed in previous analysis.
Furthermore, as the importance first-time retail investors place on management
and financial categories rises, since social signals increase at a slower rate, their
relative importance is reduced. In other words, social signals are more influential
to investors who place low importance on managerial or financial signals.
4.3 Summary The consistent findings amongst both qualitative and quantitative methods
suggest that although social signals always feature in first-time retail investors’
decision-making process, their importance is outweighed by other signal
categories, implying that investors are ‘independent amateurs’ (Mollick, 2013)
and act as part of a ‘collective intelligence’ (Schwienbacher & Larralde, 2010).
This implies that as equity-crowdfunding markets grow they become better at
harnessing ‘The Wisdom Of Crowds’ instead of herding and becoming less
efficient as observed by Burtch (2011).
43
The number of quality signals per category indicates that the sample interprets a
large number of social signals, second only to management signals. However,
they are of lowest importance when compared to other categories to a
statistically significant effect. Consequently, the behaviour and opinions of others
had the least influence on the sample’s perception of a venture’s quality.
Moderate positive correlations between social signals and all other categories
suggest that risk-averse investors place higher importance on social signals. The
absence of negative correlations also indicates that social signals are always
considered in the decision-making process, regardless of the importance placed
on the other categories. This implies that first-time retail investors place
importance on social signals but these are considered complementary to other
signalling categories suggesting that they exercise their own judgement when
evaluating a project’s quality.
These findings were supported with a standard multiple regression which found
that importance placed on management signals is the greatest predictor of the
mean importance of social signals, whilst financial signals were also found to be
statistically significant. Given that, the importance of social signals is reduced as
investors place more importance on management and financial categories, social
signals appear more influential to investors who place low importance on
managerial or financial signals. Consequently, investors sensitive to herding
behaviour could potentially be identified depending on the relative importance
they place on other signals.
Interestingly, females appear more risk-averse across all categories, with a
statistically significant difference in the greater importance they place on social
signals. This could have future implications for signalling and herding theorists,
as it appears females are more susceptible to, and place more importance on, the
opinions of others. Although when analysed through a multiple regression, it was
not found to be a significant predictor of the mean importance placed on social
signals. This study, therefore, highlights the necessity for future research to
explore a potential difference in gender risk aversion and interpretation of social
signals in early-stage venture valuation and investment.
44
These quantitative findings were consistent with the interviews where; although
some signals are interpreted ambiguously namely; largest investment and
diversity of management team, the sample was highly consistent in their
interpretation of quality signals. Management trust was presented as a
significant signal conveyed through the entrepreneur’s personality and
preparedness, which is gained predominantly through the quality and content of
the pitch video. Moderate financial returns are of considerable importance,
whether high-returns are considered irrelevant cementing Scheder & Arboll’s
(2014) view of the role of equity-crowdfunding in entrepreneurial finance.
Interestingly browsing behaviour suggests that popular projects are most likely
to be seen and could instigate quality signals being magnified by a Matthew
Effect (Merton, 1968).
It is, however, worth noting that the two methods were slightly inconsistent with
the importance of own due diligence (M=2.73, SD=1.01 from Appendix 14), with
interviewees suggesting it is essential and questionnaire respondents suggesting
it tends towards important. Although this could be attributed to the interviews
smaller sample size and the different risk aversion of investors due to their trust
in the platform’s due diligence process.
Consequently, these findings imply that first-time retail equity-crowdfunding
investors interpret signals similarly to VCs and BAs (MacMillan et al., 1985) in
that the management and entrepreneurial signals are the most important,
however, financial returns are of less importance in the equity-crowdfunding
context. The findings were consistent with Crowdcube’s experiences (T. Panesar,
personal communication, April 19, 2016) and contemporary research suggesting
that first-time retail investors conduct their own due diligence before investment
(Guenther et al., 2015), are reasonably accurate in their interpretation when
compared to sophisticated investors (Kim & Viswanathan, 2016), explaining why
social capital has little impact on funding success (Ahlers et al., 2015).
This suggests that equity-crowdfunding has multiple societal benefits as
platforms utilise ‘The Wisdom Of Crowds’ to effectively evaluate project’s value
and select high-quality projects. This appears to be the case amongst emerging
45
research (Nabarro & Altfi, 2015), although it is still too new a phenomenon for
the findings to be conclusive.
5. Limitations One principal limitation to this study is that the classification of only a few
signalling categories, primarily social signals, does not provide a comprehensive
depiction of first-time retail investors’ due diligence process. Therefore, the
study is exploratory and so was concerned in observing and testing behaviours
observed within the sample and not intended to infer the population’s
behaviour.
Another considerable limitation is the sample size as conducting more
interviews would have elicited a more comprehensive list of quality signals and a
higher number of questionnaire respondents would have increased the accuracy
of the findings.
Furthermore, it is important to note interviewees were not considering
investment when eliciting quality signals, and so their interpretation of signals
may have been biased. Additionally, as explained by Walsham (1995), since the
collection and analysis of data involves the researcher's own subjectivity, the
results should be taken with caution.
6. Research Implications As an exploratory study the principle value of this paper is in its ability to
describe the observed behaviour within the sample in order to provide a basis
for future research hypotheses. However, the findings also provide potential
implications for legislators and practitioners.
5.1 Legislation Although current legislation limits retail investors’ investment amounts, given
that they interpret signals similarly to professional investors, focus should
46
instead be directed on platforms to emit unambiguous signals for interpretation.
Consequently, signals regarding the largest investment could be clarified and
practices suggesting platforms manipulate and remove potential investors
comments (Hurley, 2016) should be investigated. Through providing clearer
signals, it follows that higher quality ventures would be selected which in turn
would reduce retail investors’ losses.
5.2 Practitioners Platforms should seek to be as transparent as possible in regards to the signals
provided on project pitches to support investors’ due diligence process. Similar
research could allow platforms and entrepreneurs to increase the amount of
credible signals on their project pitches in order to enhance investor decision-
making; leading to more successfully funded projects, increased revenue and
enriched brand reputation.
From a strategic management perspective, given that females are more risk-
averse and place higher importance on social signals, platforms could tailor pitch
information by investors’ gender to encourage investment and increase female
participation. However, effective communication remains a paramount part of
the signalling process and so platforms should ensure project information is
presented in a balanced manner so as not to obscure credible signals with those
considered most important by investors.
5.3 Future Research This study suggests that the importance first-time retail investors place on social
signals varies according to investor characteristics including the importance they
place on both management and financial signal categories. Additionally, females
were found to be more risk-averse in their due diligence process, to a statistically
significant effect with regards to social signals. Future research could be
conducted into gender differences to determine their predictive ability into the
extent of social influence in early-stage venture valuation and investment
contexts. The observed gender difference should be explored amongst signalling
47
and herding theorists, as it appears females are more susceptible to and place
more importance on social signals.
Utilising the method and categories proposed in this study researchers could
begin to categorise individual quality signals consistently to better understand
the most important signals and provide a firmer basis for comparison
throughout other crowdfunding contexts. Furthermore, the success of equity-
crowdfunding projects could be tracked to allow researchers to determine the
extent investors successfully utilise ‘The Wisdom Of Crowds’ relative to other
sources of finance. These findings could potentially allow researchers to extend
and test existing theory to predict successful ventures using an actuarial model
based upon signals presented in an equity-crowdfunding pitch. However, this
method also has considerable potential applications for other areas, including
marketing, where researchers need to develop, test and compare unknown
variables to determine their significance.
7. Conclusion To conclude, although first-time retail investors always consider social signals
when evaluating new proposals their relative importance is lower to other
signalling categories. This depicts that social signals are interpreted
complementary to other signals, suggesting that first-time retail investors use
insights from previous investors to form a ‘collective intelligence’ to evaluate
new ventures. Additionally, social signals lack of relative importance, implies
that the sample acted independently as ‘independent amateurs’ in their due-
diligence process. However, investor characteristics such as gender and
importance placed on management and financial signals have implications on the
importance of social signals, suggesting that certain portions of the population
may be more susceptible to herding behaviour. Consequently, it appears that
first-time retail investors utilise the ‘The Wisdom Of Crowds’ as oppose to
herding together, providing encouragement that as equity-crowdfunding
markets grow they will become more effective at selecting high-quality,
successful ventures.
48
8. References Abrahams, G. (2016, February 2). Why do experienced and intelligent groups make poor decisions? Speech presented at University of Exeter in Exeter. Agrawal, A., C. Catalini, and A. Goldfarb (2011): “The Geography of Crowdfund- ing,” NBER Working Paper No. 16820, http://www.nber.org/papers/w16820. Agrawal, A., Catalini, C., & Goldfarb, A. (2014). Some Simple Economics of Crowdfunding. Innovation Policy and the Economy, 14(1), 63-97. doi:10.1086/674021 Ahlers, G. K., Cumming, D., Günther, C., & Schweizer, D. (2015). Signalling in Equity Crowdfunding. Entrepreneurship Theory and Practice, 39(4), 955-980. doi:10.1111/etap.12157 Akerlof, G. A. (1970). The Market for "Lemons": Quality Uncertainty and the Market Mechanism. The Quarterly Journal of Economics, 84(3), 488. doi:10.2307/1879431 Banerjee, A. V. (1992). A simple model of herd behavior. The Quarterly Journal of Economics, 797-817. Barter, C., & Renold, E. (1999). The use of vignettes in qualitative research. Social research update, 25(9), 1-6. Baum, J. A., & Silverman, B. S. (2004). Picking winners or building them? Alliance, intellectual, and human capital as selection criteria in venture financing and performance of biotechnology startups. Journal of Business Venturing, 19(3), 411-436. doi:10.1016/s0883-9026(03)00038-7 Belleflamme, P., Lambert, T., & Schwienbacher, A. (2014). Crowdfunding: Tapping the right crowd. Journal of Business Venturing, 29(5), 585-609. doi:10.1016/j.jbusvent.2013.07.003 Belleflamme, P., Omrani, N., & Peitz, M. (2015). The economics of crowdfunding platforms. Information Economics and Policy, 33, 11-28. Benkler, Y. (2006). The Wealth Of Networks: How Social Production Transforms Markets And Freedom. New Haven: Yale University Press. Bienz, C., & Walz, U. (2006). Evolution Of Decision And Control Rights In Venture Capital Contracts: An Empirical Analysis. Available at SSRN 966155. Bolumole, Y. A., Calantone, R. J., Di Benedetto, C. A., & Melnyk, S. A. (2015). New product development in new ventures: the quest for resources. International Journal of Production Research, 53(8), 2506-2523.
49
Bradford, C. S. (2012). Crowdfunding and the federal securities laws. Columbia Business Law Review, 2012(1). Brown, R., Mawson, S. &Mason, C. (2014). Increasing ‘The Vital 6 Percent’: Designing Effective Public Policy to Support High Growth Firms. Nesta Working Paper 14/01. Retrieved March 9, 2014 from www.nesta.org.uk/wp14-01 Bryman, A., & Bell, E. (2015). Business research methods. Oxford University Press, USA. Burtch, G. (2011). Herding Behavior as a Network Externality. ICIS 2011 Proceedings, Paper 28. Retrieved April 14, 2016, from http://aisel.aisnet.org/icis2011/proceedings/humanbehavior/28/ Burtch, G., Ghose, A., & Wattal, S. (2013). An Empirical Examination Of The Antecedents And Consequences Of Contribution Patterns In Crowd-Funded Markets. Information Systems Research, 24(3), 499-519. Busenitz, L. W., Fiet, J. O., & Moesel, D. D. (2005). Signaling in Venture Capitalist—New Venture Team Funding Decisions: Does It Indicate Long‐Term Venture Outcomes?. Entrepreneurship Theory and Practice, 29(1), 1-12. Cabral, L. (2012). Reputation on the Internet. The Oxford Handbook of the Digital Economy, 343-354. Cardon, M. S., Wincent, J., Singh, J., & Drnovsek, M. (2009). The nature and experience of entrepreneurial passion. Academy of Management Review,34(3), 511-532. Carree, M. A. and R. Thurik, (2003). The Impact of Entrepreneurship on Economic Growth, in David B. Audretsch and Zoltan J. Acs (eds.). Handbook of Entrepreneurship Research, Boston/Dordrecht:Kluwer- Academic Publishers, 437–471. Chen, X. P., Yao, X., & Kotha, S. (2009). Entrepreneur passion and preparedness in business plan presentations: a persuasion analysis of venture capitalists' funding decisions. Academy of Management Journal,52(1), 199-214. Collins, L., & Pierrakis, Y. (2012, July). The Venture Crowd - Crowdfunding Equity Investment into Business. Retrieved April 15, 2016, from https://www.nesta.org.uk/sites/default/files/the_venture_crowd.pdf
Connelly, B. L., Certo, S. T., Ireland, R. D., & Reutzel, C. R. (2011). Signalling Theory: A Review And Assessment. Journal of Management, 37(1), 39-67. Creswell, J. W. (2013). Research design: Qualitative, quantitative, and mixed methods approaches. Sage publications.
50
Crosetto, P., & Regner, T. (2014). Crowdfunding: Determinants of success and funding dynamics (No. 2014-035). Jena Economic Research Papers. Crowdcube. (2016). The story and stats behind Crowdcube's success so far. Retrieved April 18, 2016, from https://www.crowdcube.com/infographic Crowdfund Insider. (n.d.). The Ultimate Crowdfunding Guide - Crowdfund Insider. Retrieved April 15, 2016, from http://www.crowdfundinsider.com/the-ultimate-crowdfunding-guide/ CrowdRating. (2016). CrowdRating: The ratings agency for equity crowdfunding. Retrieved April 24, 2016, from http://www.crowdrating.co.uk/ Crowdsurfer. (2016). Campaign metrics by platform. Retrieved April 18, 2016, from https://www.crowdsurfer.com/app/#!/analysis/campaigns/byplatform Daymon, C., & Holloway, I. (2010). Qualitative research methods in public relations and marketing communications. Routledge. Eagly, A. H. (2013). Sex differences in social behavior: A social-role interpretation. Psychology Press. Edmondson, A. C., & McManus, S. E. (2007). Methodological fit in management field research. Academy of management review, 32(4), 1246-1264. Elfenbein, D. W., Fisman, R., & McManus, B. (2012). Charity as a substitute for reputation: Evidence from an online marketplace. The Review of Economic Studies, 79(4), 1441-1468. FCA. (2013, June). Restrictions on the retail distribution of unregulated collective investment schemes and close substitutes. Retrieved April 15, 2016, from https://www.fca.org.uk/static/documents/policy-statements/ps13-03.pdf FCA. (2015). A Review Of The Regulatory Regime For Crowdfunding And The Promotion Of Non-Readily Realisable Securities By Other Media. Retrieved April 14, 2016, from https://www.fca.org.uk/static/documents/crowdfunding-review.pdf Franke, N., Gruber, M., Harhoff, D., & Henkel, J. (2008). Venture Capitalists' Evaluations of Start‐Up Teams: Trade‐Offs, Knock‐Out Criteria, and the Impact of VC Experience. Entrepreneurship Theory and Practice, 32(3), 459-483. Frydrych, D., Bock, A. J., Kinder, T., & Koeck, B. (2014). Exploring Entrepreneurial Legitimacy In Reward-Based Crowdfunding. Venture Capital,16(3), 247-269. Gerber, E. M., & Hui, J. (2013). Crowdfunding. ACM Transactions on Computer-Human Interaction ACM Trans. Comput.-Hum. Interact. TOCHI, 20(6), 1-32. doi:10.1145/2530540
51
Gerber, E. M., Hui, J. S., & Kuo, P. Y. (2012, February). Crowdfunding: Why People Are Motivated To Post And Fund Projects On Crowdfunding Platforms. In Proceedings of the International Workshop on Design, Influence, and Social Technologies: Techniques, Impacts and Ethics. Giudici, G., Guerini, M., & Rossi Lamastra, C. (2013). Why Crowdfunding Projects Can Succeed: The Role Of Proponents’ Individual And Territorial Social Capital. SSRN Electronic Journal. Available at: http://ssrn.com/abstract=2255944 Guenther, C., Hienerth, C., & Riar, F. (2015). The Due Diligence of Crowdinvestors: Thorough Evaluation or Gut Feeling only? Academy of Management Proceedings, 2015(1), 16862-16862. doi:10.5465/ambpp.2015.16862abstract Guest, G., Bunce, A., & Johnson, L. (2006). How Many Interviews Are Enough? An Experiment With Data Saturation And Variability. Field methods,18(1), 59-82. Hemer, J. (2011). A snapshot on crowdfunding (No. R2/2011). Working papers firms and region. Henrekson, M. &Johansson, D. (2010) Gazelles as job creators: a survey and interpretation of the evidence, Small Business Economics, 35(2), 227-244. Herzenstein, M., Dholakia, U. M., & Andrews, R. L. (2011). Strategic herding behavior in peer-to-peer loan auctions. Journal of Interactive Marketing,25(1), 27-36. Hoang, H., & Antoncic, B. (2003). Network-based research in entrepreneurship: A critical review. Journal of business venturing, 18(2), 165-187. Hornuf, L., & Schwienbacher, A. (2014). Crowdinvesting–Angel Investing for the masses?. Handbook of research on venture capital, 3. Howe, J. (2006, June 1). The Rise of Crowdsourcing. Retrieved April 14, 2016, from http://www.wired.com/2006/06/crowds/ Hsu, D. H. (2007). Experienced entrepreneurial founders, organizational capital, and venture capital funding. Research Policy, 36(5), 722-741. Hsu, D. H., & Ziedonis, R. H. (2013). Resources as dual sources of advantage: Implications for valuing entrepreneurial‐firm patents. Strategic Management Journal, 34(7), 761-781. Hurley, J. (2016, February 9). Crowdcube 'uses neat trick to fool investors' | The Times. Retrieved April 18, 2016, from http://www.thetimes.co.uk/tto/money/article4686065.ece Jin, G. Z., & Kato, A. (2007). Dividing online and offline: A case study. The Review of Economic Studies, 74(3), 981-1004.
52
Keuschnigg, C. (2004). Venture Capital Backed Growth. Journal of Economic Growth, 9(2), 239-261. doi:10.1023/b:joeg.0000031428.35711.fc Kim, K., & Viswanathan, S. (2016). The 'Experts' in the Crowd: The Role of 'Expert' Investors in a Crowdfunding Market. SSRN Electronic Journal. Available at: http://ssrn.com/abstract=2258243. Kuppuswamy, V., & Bayus, B. L. (2015). Crowdfunding creative ideas: The dynamics of project backers in Kickstarter. UNC Kenan-Flagler Research Paper, (2013-15). Landström, H. (1998). Informal Investors As Entrepreneurs. Technovation,18(5), 321-333. Lehner, O. M. (2013). Crowdfunding social ventures: A model and research agenda. Venture Capital, 15(4), 289-311. doi:10.1080/13691066.2013.782624 Levie, J. & Gimmon, E. (2008). Mixed signals: Why investors may misjudge first time high technology venture founders. Venture Capital, 10, 233–256.
Lewis, G. (2011). Asymmetric information, adverse selection and online disclosure: The case of eBay motors. The American Economic Review,101(4), 1535-1546. Macmillan, I. C., Siegel, R., & Narasimha, P. (1985). Criteria Used By Venture Capitalists To Evaluate New Venture Proposals. Journal of Business Venturing,1(1), 119-128. doi:10.1016/0883-9026(85)90011-4 Mason, C. M., & Harrison, R. T. (2002). Is it worth it? The rates of return from informal venture capital investments. Journal of Business Venturing, 17(3), 211-236. Mason, C., & Stark, M. (2004). What do investors look for in a business plan? A comparison of the investment criteria of bankers, venture capitalists and business angels. International Small Business Journal, 22(3), 227-248. Mass Solution. (2015, April 07). 2015CF Crowdfunding Industry Report. Retrieved April 14, 2016, from http://www.crowdsourcing.org/editorial/global-crowdfunding-market-to-reach-344b-in-2015-predicts-massolutions-2015cf-industry-report/45376 Masters, T. (2013). Marillion 'understood where the internet was going early on' Retrieved April 14, 2016, from http://www.bbc.co.uk/news/entertainment-arts-23881382 Merton, R. K. (1968). The Matthew effect in science. Science, 159(3810), 56-63.
53
Mitra, D. (2012). The role of crowdfunding in entrepreneurial finance. Delhi Business Review, 13(2), 67. Mollick, E. R. (2013). Swept away by the crowd? Crowdfunding, venture capital, and the selection of entrepreneurs. Venture Capital, and the Selection of Entrepreneurs (March 25, 2013). Mollick, E. (2014). The Dynamics Of Crowdfunding: An Exploratory Study. Journal of business venturing, 29(1), 1-16. Nabarro & Altfi Data. (2015, November). Where Are They Now? - A report into the status of companies that have raised finance using Equity Crowdfunding in the UK. Retrieved April 24, 2016, from http://www.altfi.com/downloads/WhereAreTheyNow.pdf ONS. (2016). 2015 Gross Annual Household Savings - Office for National Statistics. Retrieved April 13, 2016, from http://www.ons.gov.uk/economy/grossdomesticproductgdp/timeseries/nssh Orlikowski, W. J. (1993). CASE tools as organizational change: Investigating incremental and radical changes in systems development. MIS quarterly, 309-340. Parsont, J. W. (2013). Crowdfunding: The Real and the Illusory Exemption. Harvard Business Law Review (Spring 2014). Pricco, D. (2015). SEC's new JOBS Act Title III Crowdfunding Rules: Overview and First thoughts - CrowdExpert.com. Retrieved April 14, 2016, from http://crowdexpert.com/articles/new_jobs_act_titleiii_rules_overview_first_thoughts/ Ramser, P. (1993). Review of Decision Making in Action: Models and Methods. American Psychological Association. Razali, N. M., & Wah, Y. B. (2011). Power comparisons of shapiro-wilk, kolmogorov-smirnov, lilliefors and anderson-darling tests. Journal of Statistical Modeling and Analytics, 2(1), 21-33. Reuber, A. R., & Fischer, E. (2005). The company you keep: How young firms in different competitive contexts signal reputation through their customers. Entrepreneurship theory and practice, 29(1), 57-78. Royston, P. (1992). Approximating the Shapiro-Wilk W-test for non-normality. Statistics and Computing Stat Comput, 2(3), 117-119. doi:10.1007/bf01891203
54
Scheder, B., & Arb∅ll, C. (2014, June 4). Crowdinvestor Investment Decision-Making. A Study of Motivation, Investment Process and Criteria. Retrieved April 14, 2016, from https://www.researchgate.net/publication/262917780_Crowdinvestor_Investment_Decision-Making_A_Study_of_Motivation_Investment_Process_and_Criteria Schjoedt, L., Monsen, E., Pearson, A., Barnett, T., & Chrisman, J. J. (2013). New venture and family business teams: understanding team formation, composition, behaviors, and performance. Entrepreneurship Theory and Practice, 37(1), 1-15. Schwienbacher, A., & Larralde, B. (2010). Crowdfunding of small entrepreneurial ventures. Handbook of entrepreneurial finance, Oxford University Press, Forthcoming. Shane, S., & Cable, D. (2002). Network ties, reputation, and the financing of new ventures. Management Science, 48(3), 364-381. Smith, C. (2015). Equity Crowdfunding Models: Investor-led vs Company-led. Retrieved April 18, 2016, from https://www.syndicateroom.com/learn/overview/investor-led-versus-company-led Sonenshein, S., Herzenstein, M., & Dholakia, U. M. (2011). How accounts shape lending decisions through fostering perceived trustworthiness. Organizational Behavior and Human Decision Processes, 115(1), 69-84. Steinskog, D. J., Tjøstheim, D. B., & Kvamstø, N. G. (2007). A Cautionary Note on the Use of the Kolmogorov–Smirnov Test for Normality. Mon. Wea. Rev. Monthly Weather Review, 135(3), 1151-1157. doi:10.1175/mwr3326.1 Surowiecki, J. (2004). The wisdom of crowds: Why the many are smarter than the few and how collective wisdom shapes business, economies, societies, and nations. New York: Doubleday. Tavakol, M., & Dennick, R. (2011). Making sense of Cronbach's alpha. International journal of medical education, 2, 53. Tucker, C., & Zhang, J. (2011). How does popularity information affect choices? A field experiment. Management Science, 57(5), 828-842. Tyebjee, T. T., & Bruno, A. V. (1984). A Model of Venture Capitalist Investment Activity. Management Science, 30(9), 1051-1066. doi:10.1287/mnsc.30.9.1051 Unger, J. M., Rauch, A., Frese, M., & Rosenbusch, N. (2011). Human capital and entrepreneurial success: A meta-analytical review. Journal of Business Venturing, 26(3), 341-358. Van Selm, M., & Jankowski, N. W. (2006). Conducting online surveys. Quality and Quantity, 40(3), 435-456.
55
Visagie, I. (2011), “Venture Capital Investment Criteria, An analysis of criteria and their relative importance”, Business Mastery Project, Cass Business School, City University London. Walsham, G. (1995). Interpretive case studies in IS research: nature and method. European Journal of information systems, 4(2), 74-81. Ward, C., & Ramachandran, V. (2010, December). Crowdfunding the next hit: Microfunding online experience goods. In Workshop on Computational Social Science and the Wisdom of Crowds at NIPS2010. Yan, S. (2015). Which signalling factors facilitate the success probability of equity crowdfunding? (Unpublished master's thesis). University of Twente. Retrieved April 18, 2016, from http://essay.utwente.nl/68537/1/Yan_BA_MB.pdf Zacharakis, A. L., & Meyer, G. D. (2000). The potential of actuarial decision models: can they improve the venture capital investment decision?. Journal of Business Venturing, 15(4), 323-346. Zhang, B., Baeck, P., Bone, J., Ziegler, T., & Garvey, K. (2016, February). Pushing Boundaries The 2015 Uk Alternative Finance Industry Report. Retrieved April 13, 2016, from https://www.nesta.org.uk/sites/default/files/pushing_boundaries_0.pdf Zhang, J., and P. Liu (2012): “Rational Herding in Microloan Markets,” Management Science, 58(5), 892–912. Zimmerman, M. A. (2008). The influence of top management team heterogeneity on the capital raised through an initial public offering. Entrepreneurship Theory and Practice, 32(3), 391-414.
56
9. Appendices
Appendix 1 – Types of Crowdfunding Crowdfunding can be segmented into two broad types; donation based and investment based. Recently there has also been a rise in hybrid forms, which combine some features from all types, offering rewards in addition to equity for example (Mass Solution, 2015). Some academics, notably Scheder, & Arboll (2014) and Gunther et al. (2015) refer to investment-based crowdfunding activities as crowdinvesting in an attempt to separate the two types given the increased risk and different motives of investors. However, the terms crowdfunding and crowdinvesting will be used interchangeably within this paper since crowdinvesting has yet to gain traction amongst the relevant literature. Donation based-crowdfunding can be segmented by reward-based crowdfunding and charity-based crowdfunding: Reward-based crowdfunding currently receives the most media attention and allows entrepreneurs to compensate contributors at different reward levels with a personalised thank-you, pre-selling products or services as examples. This form is most commonly used in the form of pre-selling as market validation for an idea, product or concept. Charity-based crowdfunding refers to accepting charitable donations for a number of causes or non-profits. Investment-based crowdfunding occurs when contributors exchange money for securities and can be dissected into three segments; debt-based, equity-based and royalty-based: Debt-based crowdfunding is currently the largest form of crowdfunding and allows contributors to lend money to an individual or company with the expectation that the loan will be repaid with interest. In addition to providing loans in the UK investors have the option to purchase ‘Mini-Bonds’ which allow issuing companies to crowdfund unsecured debt. Equity-based crowdfunding, the topic of this paper, allows contributors to exchange capital for company equity, or ownership. Companies can issue different share types, which may or may not provide pre-emption or voting rights depending on the company and investment amount. Recently a distinction has been made between real-estate equity-crowdfunding and non-real-estate crowdfunding. Equity-crowdfunding can be defined as the sale of registered securities, by mostly early stage firms, to both retail, sophisticated and institutional investors. Real-estate equity-crowdfunding is a more specialist form, which involves direct investment into a property by individuals, usually through the sale of a registered security in a special purpose vehicle.
57
Royalty-based crowdfunding is a form of profit sharing which allows contributors to receive a percentage of revenue from the project or venture, once it begins to generate capital Adopted from (Crowdfund Insider, n.d.), (Mass Solution, 2015) and (Zhang et al., 2016).
Appendix 2 – Mass Solution The 2015CF – Crowdfunding Industry Report collects data on 1,250 active crowdfunding platforms to give some of the most accurate statistics on an incredibly lucrative and dynamic industry. Below is a summary of some of the key findings from the report. In 2014, the key types of crowdfunding by type were as follows:
Lending-based crowdfunding grew 223% to $11.08 billion Equity-based crowdfunding grew 182% to $1.1 billion Hybrid-based crowdfunding grew 290% to $487 million Royalty-based crowdfunding grew 336% to $273 million Donation- and Reward-based crowdfunding grew 45% and 84%
respectively In 2014, the regions of funding volume were as follows:
North America: crowdfunding volumes grew 145% to $9.46 billion Asia: crowdfunding volumes grew 320% to $3.4 billion Europe: crowdfunding volumes grew 141% to $3.26 billion South America, Oceania and Africa: crowdfunding volumes grew 167%,
59% and 101%, respectively In 2014, the lead categories share of funding volume were as follows:
Business & Entrepreneurship at 41.3% ($6.7bn) Social Causes 18.9% ($3.06bn) Films & Performing Arts 12.13% ($1.97bn) Real Estate 6.25% ($1.01bn) Music and Recording Arts 4.54% ($736m)
The report also forecasts total global crowdfunding to reach $34.4 billion in 2015. (Mass Solution, 2015)
58
Appendix 3 – JOBS Act The Jumpstart Our Business Startups Act (JOBS Act), is a U.S. law which aims to increase funding to small businesses by reducing various securities regulations and came into effect on April 5, 2012. The law has extreme implications for SMEs as it allowed accredited investors to use equity-crowding platforms. Recently, Title III was approved which allows retail investors to participate in equity-crowdfunding (Pricco, 2015).
Appendix 4 – Sources of Entrepreneurial Finance
Taken from: (Schwienbacher & Larralde, 2010)
Appendix 5 - Venture Capital and Business Angel Investment Criteria The following table provides an overview of the literature into quality signals in VC and BA capital markets. It is by no means an exhaustive list, but serves as a basis for observing which quality signals are used in early-stage venture valuation. The resulting categories product, management, financial and social were considered the most appropriate form of categorisation in the eyes of the researcher although they were derived principally by Baum & Silverman’s . (2004) classification of human, intellection and social (alliance) capital and Macmillan et al.’s (1985) classifications of; entrepreneur’s personality, entrepreneur’s experience, characteristics of the product or service, characteristics of the market or financial considerations. The insights gained from reviewing VC and BA literature were particularly useful when classifying and coding quality signals elicited from the semi-structured interviews.
59
60
Appendix 6 – Crowdfunding Investment Criteria The table on the following page provides an overview of the literature into quality signals in crowdfunding and other online marketplaces. It is by no means an exhaustive list, but serves as a basis for observing which quality signals are used in early-stage venture in the crowdfunding context. Once again, signals were categorised according to this researcher’s preferences but were derived principally from Ahlers et al.’s (2015) adopted categores of venture quality (human, social and intellectual capital) and level of uncertainty (equity share and financial projections). Categories were kept constant when evaluating VC, BA and crowdfunding literature and conducting the study in order to provide a basis for comparisions.
61
62
Appendix 7 - Interview Process Prior to the interview, interviewees were asked if they had ever previously invested in an equity-crowdfunding project or if they qualified as either; a sophisticated investor or high-net worth individual. If a respondent responded ‘no’ to all the above they were classified as a first-time retail investor and were given the following information to provide a context to equity-crowdfunding: “Please read the following, if you have any questions feel free to ask:
Crowdfunding is a relatively new method of early-stage financing for young businesses and entrepreneurs, where all stages of the investment process are conducted online
Entrepreneurs seeking funding, create a pitch where they describe their product or service and future business ambitions
Investors are then given the opportunity to interact with the
entrepreneurs and other investors to get information, ask questions and give feedback on the project
Investors can also financially contribute to the project by setting their
desired amount
Crowdfunded projects typically have a lot of investors (the crowd), who contribute small amounts until the funding goal is achieved
Once the funding goal has been achieved, successful projects receive their
funding and start their project
Projects typically adopt an all-or-nothing approach and so if a project does not reach its target any amount pledged by the investors is returned free-of-charge
There are many different types of crowdfunding, but the focus of this
survey is equity-crowdfunding
Equity-crowdfunding provides investors with shares in the project they invest in. This allows investors to not only financially support projects but also receive shares in the future profits of the business and, in some cases, rights to vote on company decisions.”
Before proceeding, the interviewees were asked if they were comfortable with the equity-crowdfunding process and, where applicable, their queries were discussed with the interviewer until they clearly understood the concept. The interview was conducted in three steps. To begin, interviewees were placed in front of a laptop and were directed to the crowdcube.com homepage. They were asked to browse current investment opportunities until they found one
63
they were interested in. The interviewer observed and noted their browsing behaviour and then asked the interviewees to analyse all the components of the investment opportunity’s pitch including; the promotional video, the description, team, financial and product information. Tape recording then began and the interviewees were asked to identify and locate which aspects of the investment opportunity indicated the project’s quality. The interviewer then asked a number of open-ended prompt questions to elaborate on the interviewee’s perceptions of quality features within the pitch. These questions were selected based primarily on insights from previous interviews; however, general themes from VC, BA and crowdfunding literature were also integrated. If the interviewee had already mentioned the questions they were excluded. The final list of questions were the following:
1. How did you choose this project? / Which aspects were important? 2. What do you think of equity crowdfunding as a concept? 3. How would you evaluate the quality of this project? 4. Is the company’s branding important? 5. Is the entrepreneur’s personality important?
a. Is it important that they are trustworthy? b. Is it important that they are likeable?
6. Is interaction with the entrepreneur important? 7. Is the project team important?
a. Does it have to be diverse? b. Is it important that they have experience?
8. Is the amount of information available important? a. Would you check this information?
9. Is it important to like the product? a. Do you have to understand it? b. Does it have to be new or something you are familiar with? c. Do you have to want it?
10. Are financial returns important? 11. Are other people’s views important?
a. An industry expert? b. Experienced investors? c. Celebrities?
12. Is it important to monitor the project’s progress? Interviewees chose a number of different projects out of the possible 30 current investment opportunities at the time of study (8 different projects were chosen out of 16 interviews) and all of the prompts were answered slightly differently, indicating that interviewees were not primed to respond in a particular way.
64
Appendix 8: List of Elicited Quality Signals From Interviews The axial and open codes (Appendix 9) gained from the interviews were used in conjunction with previous VC, BA and crowdfunding literature to phrase the following signals. Where applicable, the phrasing of questions was kept constant with previous research, particularly with MacMillan et al.’s (1985) study, in order to provide more reliable comparisons.
65
Appendix 9 - Example Interview Transcript For respondents who opted for confidentiality only their respondent number, gender and project choice was recorded. Additional permission was gained from the following respondent to include the transcript in this study. Respondent: 14 Name: Tash Parker Age: 21 Gender: Female Project: Renovagen Time: 17:00 – 17:40 Date: 24/03/16 Location: Exeter Browsing Behaviour:
Did not change filter for last investment Filter by sectors (leisure and tourism, environment, ethical and socially
beneficial, media and creative services) Quick scrolling through project photos and description Ignoring projects that are overfunding (“This one hasn’t reached its target
amount yet but the other one has”) Semi-Structured Interview: I: How did you choose this project? / Which aspects were important? I went through the different sectors and chose the ones that I knew I was interested in and I have experience in (leisure and tourism, environment, media and creative services). The only ones that really matched my interest were socially beneficial projects. One of the projects looked really good, and had more uses and potential but it had already reached it target. Whereas this other project hadn’t and it’s better to have more companies out there than not. They also had less days left to reach their target. It also has a personal connection, as it’s a product that I would find useful in my home country.
I: How would you evaluate the quality of this project? I think they did really well bringing in people that seemed authoritative in some way. To say something about why the product would be important and the humanitarian examples for the potential uses (eg. Medical camp). Obviously it is a profit-making business but I liked the societal focus.
66
The video was documentary like, but had good footage. I mean if you are going to put money into this idea then you are going to want a professional, informative video like this. It was really obviously legitimate which built my trust in the project. The way that they showed all their achievements to date and a clear plan meant that they really laid it out well. I didn’t really get the gist of how they were going to achieve the plan so there were a few details missing but otherwise good. I also liked the Cambridge university degree graduate, that immediately ups the legitimacy and status almost of the people, which was clever. I: Is the entrepreneur’s personality important? I think it shouldn’t be, but if they are going to be on camera then you have to understand that people will make decision based on what they see and hear. So if you maybe know that you don’t come across well on camera then you should find another way to get the idea across. In this example the guy was obviously genuine but didn’t come across very excited which may mean that people will lose interest, if they aren’t emotionally connected to him. Even if you have an amazing product, you need to sell it in a way that makes people feel part of it and want to spend time, money and risk with. So yea, the personality is important. I: Is it important that they are trustworthy? Definitely, it is very difficult to verify. I mean there are all these internet scams but I suppose the website is the security in a way so that helps with the legitimacy. You don’t want to put money into something that you’re not going to get money back. So the appearance of trustworthiness is important, but it is something that can so easily be manipulated. However, talking about the history and the progress the company has made, not just from the company itself but all the experts and CEOs of top customers make it seems trustworthy. The fact that they were tagged in the sector of charity and environment also helps. The idea that there is a small team makes it seems more trustworthy, because they have dedicated a lot of time and effort into the idea and are just starting up. I: Is interaction with the entrepreneur important? Not necessarily with the entrepreneur if they don’t come across well but with the team definitely. The entrepreneur might not be the main focus, but it is important to see faces for where you’re money is going rather than statistics. Especially to grab initial interest to get people involved with stories, plans, achievements and set backs etc.
67
They mention that they’ve filed patents and you want to keep track of how they get on with that given that it’s such a good product. I: Is the project team important? You want to know that other people are passionate about the same product and have faith in it. So there is that relative crowd mentality where you see people that seem like they are relatively successful, intelligent people that are willing to spend their time focussing on this idea. Gives you a small amount of faith that they believe in it and so there is more of a chance of success. I: Does it have to be diverse? Yea I think that it is important. It’s the same with anything, once you have the idea you want to appeal to a wide audience so you want people on the team that people can relate to. They shouldn’t be picked for diversity just because of the ethical aspects of it, but having a diverse team widens your chance of investors finding a connection with the team or product. You also know that each person coming into the team will come at it with a different point of view or perspective which could be beneficial. If you have a bunch of Eton kids deciding to start a product, they will have a one-sided view of the world unless they come from different backgrounds. I: Is the amount of information provided on the pitch important? Is it information that you trust? I’d have to know more about the actual platform and how they do their background checks. Initially I’d say yes I trust it but obviously if I was considering putting money in then I would have to do additional checks. I wouldn’t just go off the information provided off this page, I’d find out what other people think of this company and a basic Google search. Perhaps you would contact the team and get a feel for what they are like. But yea it looks trustworthy. I: Does the product have to be new or something you are familiar with? It depends, I wouldn’t want to put money into something that I thought would be something that already exists. It would have to be something new, and it would have to be something I believe in as a concept or ideology (eg. Charity). That doesn’t necessarily mean new but it does have to have new enough elements that I have the feeling and belief that it’s going to catch on. I: Are financial returns important? Does crowdfunding differ from charity? I mean you’ve obviously got some idea that you are going to get some money back, but again that would depend on how much you’re going to put in. You are still going to take a risk, but at least you know that risk is in hope of something
68
good which is not just for yourself. You’re not just buying lottery tickets, I mean even if it’s not going to give you massive returns hopefully it will be something that triggers positive change. So your money has been used well even if it doesn’t come back. I’d say it’s probably more an intrinsic motivation than an extrinsic one.
I: You mention that it is important that other people are passionate about the product but how important are other people’s views when considering the quality of the project? Everything is going to come up against problems so you want the team to be passionate enough to overcome those difficulties. I did have a look at the number of investors, and noticed there were a lot of investors and awards, experts and customers that like the project which improved my confidence. However, other peoples views weren’t the biggest thing and it was the concept that was the main decisive factor. I: Experienced investors? The largest investment didn’t affect my decision at all to be honest I didn’t even see it. However, if there was a venture capitalist on board I would think well of the project but at the same time I wouldn’t expect that the project had secured an experienced investor. I: Why would it make you think well of the project? I mean whenever you hear of an expert making a decision it affects your judgement because you expect them to know more about it then you do. But at the same time it doesn’t mean you are going to make a decision based on what they do but you get a feeling that what you are doing is good, in comparison to other choices.
Appendix 10 - Open and Axial Codes The following page provides a summary of the open and axial codes used for qualitative analysis when ranked by the number of references (in brackets). Codes were also used to identify interviewer sentiment when discussing a particular signal to allow for easier comparisons. For example largest investment positive and largest investment negative were considered two separate codes. The number of references is a measure of significance since it represents the most commonly recognised and discussed quality signals.
69
70
The following screenshot provides an overview of the NVivo software when the code of ‘High Largest Investment as Ambiguous’ was selected. These codes allowed all relevant quotations to be compared easily to give the sample’s consensus on a particular signal or area of interest.
If a quotation was found to be of importance, it would be highlighted in the interview transcripts for further investigation and to find the respondent’s number:
Appendix 11 - Online Questionnaire The following shows screenshots of the online questionnaire. The online survey software Surveygizmo.com was used to collect responses which were distributed on social media and through iPads passed to willing participants.
71
72
73
74
75
76
77
78
79
Appendix 12 - Data Cleaning A combination of techniques were used to clean the data to monitor for respondents who were; straight-lining, using Christmas-tree behaviour, giving gibberish answers, speeding and being inconsistent. Straight-lining refers to when a respondent answers the same option for each item without reading the question, and can be easily detected when reviewing data. There were no straight-lining responses detected in the questionnaire responses. Christmas-tree behaviour refers to when the respondent answers questions in a particular pattern (ie. 1-2-3-4-3-2-1). Surveygizmo uses an algorithm to detect this behaviour, however, none were detected in this survey. Gibberish answers refer to incoherent answers and are flagged as being spelling mistakes. This was used to detect whether respondents had correctly entered their age as opposed to a random sequence of letters. Again, none were detected in the survey. Another, more frequent, habit of online survey respondents is to speed through answers without accurately considering the questions. Consequently, given the length of the survey with 73 questions some responses were excluded based on the amount of time they spent on question pages. The slowest 7 % of responses were excluded from calculating the average time per question, as a result of interpreting the graph, to make allowances for respondents with reading difficulties. The resulting average time per question was then used to detect extremely quick answers. In combination with reviewing the graph and considering the unlikelihood that respondents could accurately respond to questions in anything less than 4 seconds, those answers were removed. Consequently, the fastest 15% of responses (a total of 22) were removed from the data (shown below). It is important to note that out of the responses removed, only 6 were completed questionnaires, whereas the others referred to partial responses which were used for testing.
80
A final check was conducted with regards to the overall consistency of respondents’ answers. Consequently, a small number of questions were selected on their ability to show contradictory responses. Question logic was then constructed so that if a respondent made an extreme contradictory response it was flagged for monitoring. The 6 contradictory response cases incorporated into the question logic were the following: Case 1: 10. The product has an existing customer base. Irrelevant 16. The venture is already an established business. Essential or 10. The product has an existing customer base. Essential 16. The venture is already an established business. Irrelevant Case 2: 35. The team update their project regularly. Irrelevant 36. The team respond quickly and insightfully to questions. Essential or 35. The team update their project regularly. Essential 36. The team respond quickly and insightfully to questions. Irrelevant Case 3: 46. I require a return at least equal to my investment. Irrelevant 47. I require at least a competitive return on my investment. Essential Case 4: 46. I require a return at least equal to my investment. Irrelevant 48. I require at least 10 times return on my investment. Essential Case 5: 47. I require at least a competitive return on my investment. Irrelevant 48. I require at least 10 times return on my investment. Essential Case 6: 65. The venture has a social media presence. Irrelevant 66. The venture has a large number of social media followers. Essential There were, however, no contradictory responses found in the survey. To summarise, of the 179 survey responses, 42 were partially completed, 12 were disqualified according to their investor classification and 22 were removed for speeding resulting in 103 valid, complete responses.
81
Appendix 13 – Reliability Checks
Cronbach Alpha The Cronbach Alpha test is commonly used in social sciences to estimate data reliability by testing for internal consistency. Internal consistency describes the inter-relatedness of test items by measuring the extent to which all the variables in a test measure the same concept or construct. It is an appropriate measure of reliability in this case as the test is reasonably long (58 variables) and since the Likert-scale only measures one concept. Generally it is considered that a score of greater that 0.7 provides an acceptable level of reliability, with a score of 0.8 implying good reliability and a score of .9 excellent reliability. Consequently, this study’s score of .86 suggests good reliability, as shown below:
Shapiro-Wilk Test The Shapiro-Wilk test is used to test whether a sample came from a normally distributed population. The assumption that data follows a normal distribution is essential for parametric testing used in this study namely; independent and pair samples T-tests and multivariate regressions. The results of the normality tests are shown below:
The results of Kolmogorov-Smirnov can be ignored as it is best suited to large samples (Royston, 1992) and its accuracy has recently been called into question (Steinskog et al., 2007). Additionally the Shapiro-Wilk test has been proven to be more accurate and powerful at smaller sample sizes of n=100 (Razali et al., 2011), making it highly applicable to this study. Due to the results of the Shapiro-Wilk test, at the 0.05 significance level, the hypothesis that the sample comes from a population which has a normal
82
distribution cannot be rejected. This suggests parametric statistical analysis will provide accurate results when used in this study. Interestingly, the financial category is close to being classified as a non-normal distribution. Future research should attempt to clarify whether this is consistent amongst larger sample sizes, which would suggest that investors’ interpretations of financial signals are inherently skewed. Graphical results, such as a quartile-quartile plot and histogram analysis reflect these findings as shown below:
The histogram analyses on the following pages also reflect that the categories follow a normal distribution, although minor abnormalities can be seen with regards to financial signals.
83
84
85
Appendix 14 – Removed Quality Signals
Appendix 15 - MacMillan et al.’s (1985) Result Comparison The following two figures demonstrate that the sample was very consistent with the importance of entrepreneur and management both in terms of the criteria most frequently rated essential and in the mean importance placed on each signal. The comparison also reflects the different role equity-crowdfunding has for entrepreneurial finance as there was more of a focus on newer projects, with the sample placing a lower mean importance on previous track record and industry experience, and a significantly lower importance on high financial returns.
86
The differences in mean importance for individual signals can be shown in the table below:
Appendix 16 – Top 20 Highest Ranked Variables
87
Appendix 17 – Independent-Samples T-Test and Age Groups Ages were grouped in the following categories; 18-24, 25-34, 35-44, 55-64, 65+ in order to provide a firmer basis for comparison. However, given that the sample had a very homogenous age range, only the two largest age ranges were used for the Independent-Samples T-Test (18-24 and 25-34). However, this was not found to be a statistically significant predictor of any of the signal categories.
Appendix 18 – Ethics Form Please find the ethics form in the following pages. If there are any queries or concerns the researcher can be contacted at [email protected].
88
Part A: Background of the research project
Title of Research Project
Signalling In Equity-Crowdfunding: An Exploratory Study Into Herding Behaviour Amongst First-Time Retail Investors
Name of student Luke Barratt
Module code and title BUS3001 Dissertation
Email contact [email protected]
Supervisor’s name Dr. Boyi Li
Name(s) of other researchers and affiliation (s) (e.g. if you are conducting the research with help from a third party or being sponsored by another organisation or supported in kind)
N/A
Start and estimated end date of project
September 2015 – April 2016
Source of any funding for the project
N/A
Aims and objectives of the project (please provide as bullet points)
Provide exploratory findings into the relative
importance first-time retail investors place on social
signals within equity-crowdfunding pitches
Principal aim is to see whether they use social signals
as complementary in their due diligence process or if it
is of highest importance
The relative weight should reflect whether they act
independently in their investment valuation or
collectively as part of a herd.
89
Part B: Ethical Assessment Please complete the following questions in relation to your research project. Many of these will not be relevant to your project, but the Research Ethics Committee wishes for you to consider the full range of questions as part of your research training.2
Section 1 Yes No
Research that may need to be reviewed by NHS NRES Committee, Ministry of Defence Research Ethics Committee (MODREC) or an external Ethics committee. See http://www.hra.nhs.uk/about-the-hra/our-committees/nres/ and https://www.gov.uk/government/groups/ministry-of-defence-research-ethics-committees for more information.
X
Will the study involve recruitment of patients or staff through the NHS or the use of NHS data or premises and/ or equipment?
X
Does the study involve participants age 16 or over who are unable to give informed consent? (e.g. people with learning disabilities: see mental Capacity Act 2005 / Adults with Incapacity (Scotland) Act 2000. All research that falls under the auspices MCA/AWI should be reviewed by a recognised and appropriate REC operating under GAfREC or Scotland ‘A’ REC).
X
Section 2
Does the research involve other vulnerable groups: children, those with cognitive impairment, or those in unequal relationships? Have you read the appropriate Act; ethical practices governing research with the group you aim to study?
X
Will the study require the co-operation of a gatekeeper for initial access to the groups or individuals to be recruited? (e.g. employees, students at school, members of self-help group, residents of a nursing home?)
X
Will it be necessary for participants to take part in the study without their knowledge and consent at the time? (e.g. covert observation of people in non-public places, use of deception in experimental studies)
X
Will the study involve discussion of sensitive or potentially sensitive topics? (e.g. sexual activity, drug use, personal lives)
X
Are drugs, placebos or other substances (e.g. food substances, vitamins) to be administered to the study participants, or will the study involve invasive, intrusive or potentially harmful procedures of any kind?
X
Will tissue samples (including blood or saliva) be obtained from participants? X
Is pain or more than mild discomfort likely to result from the study? X
Could the study induce psychological stress or anxiety or cause harm or negative consequences beyond the risks encountered in normal life?
X
Will the study involve prolonged or repetitive testing? X
Will the research involve administrative or secure data that requires permission from the appropriate authorities before use?
X
Is there a possibility that the safety of the researcher may be in question? (e.g. working alone and physically present in an unfamiliar international environment)
X
2 ESRC ethics initial checklist, Framework for Research Ethics (FRE), (2015).
90
Section 2 continued… Yes No
Does the research involve members of the public in a research capacity (participant research)?
X
Will the research take place outside the UK? X
Will the research involve respondents to the internet or other visual/ vocal methods where respondents may be identified? (e.g. through the findings)
X
Will research involve the sharing of data or confidential information beyond the initial consent given?
X
Will financial or other inducements (other than reasonable expenses and compensation for time) be offered to the participants?
X
If you have answered ‘yes’ in Section 1 of the Research checklist
Your research is likely to be subject to specific ethics review other than the
University of Exeter, therefore it is unlikely that you will have sufficient time to gain
ethical approval. External permissions can often take between 3 and 12 months to
gain approval. Therefore, we advise that you revise your research proposal. If you do
wish to go ahead, please contact your supervisor and module co-ordinator.
If you have answered ‘yes’ to any of the other questions in Section 2 of the
Research checklist
You will need to describe more fully how you plan to deal with the ethics issues
raised by your research below in Sections C, D, E, F, before obtaining signatures in
section G.
Please note that it is your responsibility to follow the University of Exeter’s Code of
Practice on Ethical Standards and any relevant academic or professional guidelines in
the conduct of your study. This includes providing appropriate information sheets
and consent forms, and ensuring confidentiality in the storage and use of data. Any
significant change in the question, design or conduct over the course of the research
should be notified to your primary supervisor and may require a new application for
ethics review.
If you have answered no to all of the questions in sections 1 and 2
Please sign the form in section G and obtain your supervisors signature.
91
Part C: Further and brief details for any sections answered ‘Yes’. If you answered ‘yes’ to any of the above sections (i.e. the checklist), please elaborate with detail here. Please state:
who is at risk: e.g. the participants; yourself; organisations you are working
with.
what type of harm the research may cause: e.g. health and safety issues;
reputational damage; distress, embarrassment, anxiety; inconvenience, time
lost, intrusion, boredom or discomfort.
How the risks will be minimised and harm limited: e.g. inform someone of
your whereabouts in case of emergency; not giving your personal details to
participants in the research; limiting the type of questions you ask
respondents; giving participants the right to withdraw from the research at
any time etc…
N/A.
92
Part D: Ethical Considerations for method(s). List each of the methods you aim to use to recruit participants and describe the methods you will use to gain their ‘informed consent’. If written consent will not be obtained for any of your methods, this must be justified. At the least the following should be considered for each method. - Confidential and anonymity for all participants and organisations - Voluntary participation following informed consent - Please attach a copy of every Information Statement and Consent Form that will
be used, including translation if research is to be conducted with non-English speakers. If consent is to be obtained verbally, please indicate the script you will use to inform the participant and the method of recording the verbal agreement.
Method Please state how you obtain informed consent…
For interviews participants will be offered confidentiality regarding their identity and participation in the study.
Interviewees will be asked if they would like their details to be confidential, if not then they will not be interviewed.
For online questionnaires, respondents were asked if they gave consent for their age and gender to be used for the purpose of the research study.
Respondents cannot continue with the survey or submit responses without given their consent
Will there be any possible harm that your project may cause to participants (e.g. psychological distress or repercussions of a legal, political or economic nature)? What precautions will be taken to minimise the risk of harm to participants? No.
93
Part E: Data protection How will you ensure the security of the data collected? What will happen to the data at the end of the project, (if retained, where and how long for). Please follow guidelines provided by the University of Exeter on Data protection to complete this section http://www.exeter.ac.uk/recordsmanagement/.
The data collected only contains general information about the respondents including age and gender. The only source of data is available on the researchers’ laptop, which will be deleted upon completion of the project.
Part F: Notes and Additional Information: Please provide any additional information which may be used to assess your application in the space below.
If there are any further questions do not hesitate to contact me at [email protected]
94
Part G: Signatures:
Supervisor’s Declaration
(Form last updated 3rd November 2015 by Adrian R. Bailey)