Sei sulla pagina 1di 198
SMALL FIRM GROWTH IN A LEAST DEVELOPED COUNTRY HOW SMALL FIRM OWNERS AFFECT THE GROWTH

SMALL FIRM GROWTH IN A LEAST DEVELOPED COUNTRY

HOW SMALL FIRM OWNERS AFFECT THE GROWTH OF THEIR FIRMS IN ZAMBIA

MWANSA CHABALA

Small firm growth in a Least Developed Country

How small firm owners affect the growth of their firms in Zambia

Mwansa Chabala

Reading committee:

prof. dr. C.T.M. Elbers prof. dr. G. Romme prof. S. N. Khapova dr. I.A.W. Wakkee dr. J. Small

Chabala, Mwansa

Small firm growth in a Least Developed Country: How small firm owners affect the growth of their firms in Zambia This book is number 29 in the ABRI Dissertation series

ISBN:

978 90 361 0528 6

Front and back cover design: Mwansa Chabala Tonny Hartog, Haveka.

Front cover photo: Mwansa Chabala. The photo shows one of the construction sites for small firm that participated in the research OneWest Minerals Limited. The photos have been used with permission from the firm owner.

© Mwansa Chabala, 2018

All rights reserved. No part of this book may be reproduced or transmitted in any form by any means, electronically or mechanically, including photocopying, recording or using any information storage and retrieval system without the written permission of the author, or, when appropriate, of the publisher of the publication

Printed by HAVEKA

VRIJE UNIVERSITEIT

Small firm growth in a Least Developed Country

How small firm owners affect the growth of their firms in Zambia

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van Doctor aan de Vrije Universiteit Amsterdam, op gezag van de rector magnificus prof.dr. V. Subramaniam in het openbaar te verdedigen ten overstaan van de promotiecommissie van de School of Business and Economics op woensdag 3 oktober 2018 om 13.45 uur in de aula van de universiteit, De Boelelaan 1105

door

MwansaChabala

geboren te Chingola, Zambia

promotoren:

prof.dr. E. Masurel

copromotoren:

prof.dr. L.J. Paas dr. J.C. van Burg prof.dr. J. Lungu

Table of contents

Chapter 1

Introduction

1

1.1

Introduction and study background

2

1.2

Research setting

7

1.2.1

Least developed countries (LDCs)

7

1.2.2

Zambia’s economic history

8

1.2.3

Small and Medium Enterprises

12

1.2.4

Zambian Construction Sector

13

1.3

Introduction to the Chapters

13

1.4

Methodology

15

Chapter 2

Measuring Small Firm Growth in African Least Developed Countries: Evidence from

the Construction Sector in Zambia

18

2.1

Introduction

21

2.2

Theoretical background

23

2.2.1

Small Firm Growth

23

2.3

Qualitative study

31

2.3.1

Selection of small firm growth indicators

32

2.3.2

Operationalization of growth

35

2.4

Quantitative study

36

2.4.1

Methodology

36

2.4.2

Measures

38

2.4.3

Data analysis

39

2.4.4

Results

40

2.5

Discussion and conclusions

46

Chapter 3

Which Founders Do Extract Wealth from their Venture? The Moderating Role of Start-

up Motivation

50

4.1

Introduction

52

4.2

Theory and hypotheses

54

4.2.1

Firm growth and personal wealth

54

4.2.2

Moderating effect of entrepreneurial motivation

56

4.3

Methodology

59

4.3.2

Measures

60

4.3.3

Data analysis

65

4.4

Results

66

4.4.1

Motivation groups

67

4.4.2

Hypothesis tests

68

4.5

Discussion

71

4.6

Conclusion

74

Chapter 5

Entrepreneurial Orientation and Firm Growth in Least Developed Countries: The

Moderating Role of the Business Environment

76

5.1

Introduction

78

5.2

Theoretical framework and research hypothesis

80

5.2.1

Entrepreneurial orientation

80

5.2.2

Entrepreneurial orientation and firm performance

81

5.2.3

Moderating role of the business environment

83

5.3

Methodology

86

5.3.1

Sample and data collection

86

5.3.2

Data collection

86

5.3.3

Measures

87

5.4

Results

90

5.5

Discussion

93

5.6

Limitations to the study

95

5.7

Conclusions

96

Chapter 6

How Does Entrepreneurial Passion Affect Small Firm Growth? The Mediating Role of

Entrepreneurial Alertness

97

6.1

Introduction

99

6.2

Theory and hypotheses

102

6.2.1

Entrepreneurial passion and firm growth

103

6.2.2

Entrepreneurial passion, entrepreneurial alertness and firm growth

105

6.3

Methodology

108

6.3.1

Sample and data collection

108

6.3.2

Measures

110

6.4

Results

115

6.4.1

Correlation results

115

6.4.2

Regression results

116

6.6

Limitations and future research

121

 

Chapter 7

Conclusions

123

7.1

Conclusions

124

7.1.1

Answers to research questions in empirical studies

124

7.1.2

Answer to the main research question of the thesis

127

7.2

Contributions to literature

128

7.3

Research limitations and areas for future research

130

7.4

Policy and practical implications

131

7.5

Overall conclusion

133

List of references

134

Appendix 1. Values for NCC grade categories

159

Appendix 2. Data collection questionnaire – first wave

160

Appendix 3 – Data collection questionnaire – second wave

166

English Summary

169

Nederlandse samenvatting

174

Acknowledgements

179

ABRI Dissertation series

182

List of tables

Table 2.1. Summary of small firm growth indicators and measuring methods in entrepreneurship

research (2007 to 2013)

26

Table 2.2. Small firm growth calculation formulas

29

Table 2.3. List of experts interviewed

32

Table 2.4. Number of experts recommending a particular small firm growth indicator

33

Table 2.5. Response rates for small firm growth indicators

38

Table 2.6. Result of factor analysis for firm growth measurement scales

41

Table 2.7. Confirmatory factor analysis results for growth measurement scales

42

Table 2.8. Reliability assessment results for growth measurement scales

44

Table 2.9. T-test results comparing NCC “growth” and “no growth” Groups

45

Table 2.10. Reliability and validity results for all measurement scales

46

Table 3.1. Factor loadings and reliability results for firm start-up motivation

64

Table 3.2.Descriptive statistics and intercorrelations of variables

67

Table 3.3. Final cluster centres based on EFA factor scores

67

Table 3.4. Regression Results of relationship between personal wealth, motivation and firm growth

69

Table 3.5. Conditional Effects for motivation groups

70

Table 4.1. Factor analysis results for EO

89

Table 4.2. Correlations and descriptive statistics for firm growth, EO and business environment

 

91

Table 4.3. Regression results for relationship between EO and sales growth

92

Table 5.2. Factor analysis and reliability results for entrepreneurial alertness

112

Table 5.3. Factor analysis and reliability results for firm growth

114

Table 5.4. Factor analysis and reliability results for growth intentions

115

Table 5.5. Descriptive statistics and correlation results for passion, alertness and firm growth 116

Table 5.6. Hierarchical linear regression results for relationship between passion, alertness and

firm growth

117

Table 5.7. Indirect effects of passion for developing (via alertness) on firm growth

119

List of figures

Figure

1.1. Map of Zambia

9

Figure 3.1 Relationship between firm growth and growth in personal wealth

71

Figure 4.1. Research model and hypothesis for the relationship between EO and firm performance

86

Figure 5.1. Results of the mediating effect of alertness on the relationship between passion and

firm growth

118

List of abbreviations and acronyms

AAC

Anglo-American Corporation

AVE

Average Variance Extracted

BTS

Bartlett’s Test of Sphericity

CBU

Copperbelt University

CSO

Central Statistics Office

CFI

Comparative Fit Index

CI

Confidence Intervals

CFA

Confirmatory Factor Analysis

EA

Entrepreneurial alertness

EO

Entrepreneurial Orientation

EP

Entrepreneurial Passion

EFA

Exploratory Factor Analysis

GEM

Global Entrepreneurship Monitor

GDP

Gross Domestic Product

GNI

Gross National Income

HEART

High Education And Research Training

HDI

Human Development Index

KMO

Kaiser-Meyer-Olkin

LDC

Least Developed Countries

NCC

National Council for Construction

NUFFIC

Netherlands Universities Foundation for International Cooperation

OECD

Organisation for Economic Co-operation and Development

RST

Roan Selection Trust

RMSEA

Root Mean Square Error of Approximation

SMEs

Small and Medium Enterprises

IMF

The International Monetary Fund

TLI

Tucker Lewis Index

UN

United Nations

UNCTD

United Nations Conference on Trade and Development

UNDP

United Nations Development Programme

VIF

Variance Inflation Factors

ZANACO

Zambia National Commercial Bank

ZDA

Zambia Development Agency

ZMK

Zambian Kwacha

In loving memory of Rev. Sr. Justina Mwitwa Chabala (1963-2018)

MHSRIEP

Romans 14: 7-12

Chapter 1

Introduction

1.1

Introduction and study background

The importance of small and medium enterprises (SMEs) for economic growth has been widely

acknowledged in entrepreneurship research (Ayyagari et al., 2007; Acs et al., 2008; Box et al., 2016).

SMEs represent substantial portions of national economies in both developed and developing

countries (Soininen et al., 2011) and entrepreneurial activities of small firm owners have been

associated with economic growth (Lazear, 2005; Thurik and Wennekers, 2004; Wennekers and

Thurik, 1999; Thurik et al., 2008; Carree et al., 2007; van Stel et al., 2005). It is often argued that

economic growth is attainable through entrepreneurship which leads to competitiveness,

innovation, and employment creation through small firm formation, survival and growth (Friis et

al., 2006; Acs and Storey, 2004). To this effect, Minniti and Levesque (2010) attribute economic

growth of emerging and developing economies such as China partly to the increase in

entrepreneurial attitudes and suggest that entrepreneurship can play an equally pivotal role in

improving the economies of developing countries. Additionally, there is a general understanding

that SMEs drive economic growth through job and wealth creation (Dobbs and Hamilton, 2007).

Consequently, most developing countries employ policies aimed at promoting entrepreneurship

as a source of job and wealth creation (Cornwall and Naughton, 2003; Lazear, 2005; Naudé, 2011).

However, Shane (2009) argues that not all forms of entrepreneurship and business

formation may lead to economic growth, job and wealth creation. He labels policies that encourage

people to become entrepreneurs as “bad public policy”, on the basis that the contribution of

entrepreneurship to economic growth is not about quantity but about the quality and growth of

small firms. The quality of entrepreneurship regarding small firms in an economy can be construed

from the number of small firms which survive and grow. Small firms that grow have better chances

of survival (Andersson and Tell, 2009) and can make more significant contributions to economic

growth than those that do not (Barringer et al., 2005; Hessels et al., 2008). Similarly, the economic

benefits of entrepreneurship and business ownership may only be realised when small firms

achieve sustainable growth (Dobbs and Hamilton, 2007; Moreno and Casillas, 2007). Therefore,

2

from an economic development perspective, small firm growth is a significant aspect of the

essence of entrepreneurship (Davidsson et al., 2006). Entrepreneurs 1 who grow their firms would

to a large extent, be considered as successful and their firms are likely to contribute to national

economic growth (Davidsson et al., 2005; Barringer et al., 2005).

The realisation of the importance of firm growth in entrepreneurship has led to

numerous studies on what constitutes and determines firm growth in both developing countries

(Goedhuys and Sleuwaegen, 2010; Naudé et al., 2010) and developed countries (Davidsson et al.,

2010; Leitch et al., 2010; McKelvie and Wiklund, 2010). Although such studies have provided a

substantial body of knowledge, it is still not yet very clear what constitutes small firm growth and

how firm growth is achieved. In particular, the conceptualisation and measurement of small firm

growth has been problematic (Davidsson et al., 2013). Leitch et al. (2010) observe that not much is

known about firm growth. Kiviluoto (2013 pp. 569) further observes that firm growth research

has often been criticised as “stagnated, characterised with inconclusive research results and slow

theory development” due to lack of coherence in the way it is conceptualised and measured.

Understanding small firm growth and its antecedents has resulted in a significant debate

in entrepreneurship research (Simpson et al., 2012; Rogoff et al., 2004; Brush and Vanderwerf,

1992; Murphy et al., 1996). Firm growth has been recognised as a key dimension of business

performance (Wennekers and Thurik, 1999; Shepherd and Wiklund, 2009) resulting in research

interest in understanding differences in growth rates for small businesses. Firm growth signals high

performance and provides a positive return on investment for the business owners and is also

regarded as a way to both improve and protect the firm owner’s investment (Dobbs and Hamilton,

2007). Contrarily, lack of growth may significantly reduce the probability of survival for a small

firm (Gilbert et al., 2006).

1 In this thesis, the terms entrepreneurs, small business owners and owner-managers are used interchangeably based on Gartner’s (1989, pg. 26) definition of entrepreneurship as the “creation of new organisations”.

3

Firm growth research has focused on predicting and measuring growth, as well as gaining

cognisance of how different factors affect it (Unger et al., 2011; Song et al., 2008; Murphy et al.,

1996). Such research has produced lists of different operationalisations and determinants of firm

growth (Shepherd and Wiklund, 2009; Weinzimmer et al., 1998). The operationalisations of firm

growth that have been identified include change in number of employees, sales volume, revenue,

profitability, and size. Similarly, various factors at different levels have been found to affect firm

growth. They include factors related to the entrepreneur, the type of firms and the external

business environment. Factors in these categories have been studied individually and in different

combinations to determine how they influence business growth (Frese et al., 2002; Rauch et al.,

2009). The general conclusion that can be drawn from the body of knowledge resulting from such

research is that a variety of factors at individual, firm and environment level interact, in simple and

complex ways to influence firm growth.

The complexity of the relationship between firm growth and its antecedents has affected

the development of firm growth theories (Simpson et al., 2012; Shepherd and Wiklund, 2009). The

problem of developing firm growth theories is further compounded by the diverse nature of small

firms and idiosyncrasy of the environments in which they operate (Davidsson and Klofsten, 2003).

Therefore, it is vital to continue research in this area, especially in environments such as Least

Developed Countries (LDCs). Disparities in the business environment can be attributed to stages

in economic development of a country or region. Accordingly, a country or region’s level of

economic development can influence the drivers and obstacles of small firm growth (Okpara and

Wynn, 2007; Nichter and Goldmark, 2009; van Stel et al., 2005). It is therefore important that in

the quest to understand firm growth, economic, cultural and contextual variations between regions

are taken into consideration.

Most firm growth research has been conducted in western developed countries, making

valid generalisation of findings to other, contextually different environments, such as African

4

LDCs, less feasible. The environmental conditions faced by small firms in western developed

economies are different from those in African LDCs. Despite recognizing that the nature of

growth and its determinants would be influenced by context-specific factors (Dobbs and

Hamilton, 2007; Davidsson and Klofsten, 2003), relatively little is known about the subject in

African LDCs.

Besides being major contributors to economic growth and job creation (Acs and Mueller,

2008), small firms also have the potential to act as means for increasing personal wealth of the

owners (Cagetti and De Nardi, 2006). The role of small firms as a major source of employment

and income is even more important in LDCs (Mead and Liedholm, 1998; Unger et al., 2009). In

most African LDCs, negative environmental conditions such as unemployment and the difficulty

in finding a suitably paid job can “push” potential entrepreneurs to become de facto entrepreneurs

(Dawson and Henley, 2012; Uddin et al., 2014). Especially for such push-motivated entrepreneurs,

wealth creation is one of the primary motives to start a firm, as they do not have other (good)

options to generate sufficient income. Thus, at least for push-motivated entrepreneurs, extracting

wealth from the firm is a key motive. The rate at which the firm owners extract wealth from the

firm may affect the growth of the firm. Yet, the extent to which firm growth translates into growth

of personal wealth and how the relationship is affected by the motivation for starting a business is

still largely unknown.

It is also known that the strategic orientation of firms and their owners affects firm

growth. This demands that firms adopt a strategic orientation which would enable them to acquire

strategic entrepreneurial aspects of decision making, methods and practices (Wiklund and

Shepherd, 2005). Although research has shown that firms that adopt a more entrepreneurial

strategic orientation are relatively likely to grow (Krauss et al., 2005; Hughes and Morgan, 2007;

Lumpkin and Dess, 2001; Madsen, 2007), the exact nature of the relationship is still unknown.

This has been attributed to the differences in a firm’s entrepreneurial behaviour that are influenced

5

by context specific conditions based on the environment in which the firm operates (Lumpkin and

Dess, 2001; Miller, 2011). Entepreneurial orientation regulates the strategic decisions and resource

allocations in a firm. Therefore, understanding the effects of the strategic orientation of the

entrepreneur on firm growth in different environments is still very important.

Small firm growth is one of the important variables of entrepreneurial success (Unger et

al., 2011). Concerning entrepreneurial success, one of the central questions in entrepreneurial

research revolves around why some entrepreneurs become more successful than others (Baron,

2004a). Although part of the answer to this question is attributed to economic factors that are

external to entrepreneurial firms, the actions of entrepreneurs are also deemed relevant for small

firm growth (Shane, 2001). With such propositions, the entrepreneur, as a person, takes centre

stage as entrepreneurship research searches to better understand the role that an entrepreneur

plays in venture growth. This is despite prior research having investigated and accumulated

substantial knowledge on the effects that individual level factors, such as personality traits and

characteristics, may have on firm performance (Brandstätter, 2011; Gartner, 1989; Zhao et al.,

2009; Mitchell et al., 2002). To further enhance understanding about the role of the entrepreneur

in entrepreneurial success, entrepreneurship research has begun to focus more on the role of

entrepreneurial cognition in venture creation and success (Baron, 2004b; Shepherd, 2015; Mitchell

et al., 2002; Mitchell et al., 2007; Baron, 2000). This has led to a resurgence in research focusing on

the people-side of entrepreneurship that targets the understanding of how entrepreneurs reason,

form judgements and make decisions given the differences that exists in entrepreneurial cognitions

in order to understand while some are successful while others are not (Tang et al., 2008). However,

one key aspect of entrepreneurial cognition regarding how entrepreneurs are alert to opportunities

in the environment, and the affection that they have towards undertaking entrepreneurial activities,

still remains relatively unexplored (Baron, 2008; Grégoire et al., 2015).

6

Given these remaining questions and puzzles, this thesis investigates the measurement of

firm growth, and evaluates factors that influence firm growth in the context of an African LDC,

Zambia. The thesis sets out to contribute to answering the general question: How do small firm owners

affect the growth of their firms in LDCs? This question is relevant and important because it seeks to

identify the important dimensions and factors that can help improve the performance of small

firms in African LDCs.

1.2

Research setting

1.2.1

Least developed countries (LDCs)

This thesis focuses on small firms in an African LDC, Zambia. There are different broad

classifications of countries based on economic development used by major international

organizations. For example, the World Bank classifies countries based on Gross National Income

(GNI) per capita as low income, low middle income, upper middle income, and high income. It

uses the term “developing country” to represent low and middle-income countries although it

does not imply that all countries in this category are in the process of developing (World Bank,

2013b). The International Monetary Fund (IMF) extends this criterion to include export

diversification and the degree of integration into the global financial system. The IMF classification

divides countries into two major categories: advanced economies and emerging and developing

economies (International Monetary Fund, 2013). The United Nations Development Programme

(UNDP) criterion uses the Human Development Index (HDI), which is derived from GNI per

capita, life expectancy and educational indices. In the UNDP classification, developing countries

occupy the lower three percentiles of the distribution and countries at the lowest end of the

developing countries are termed as the Least Developed Countries.

LDCs are the poorest and weakest of the international community and are at the bottom

of all development rankings. According to the United Nations (2013), they have a low level of

socio-economic development, characterised by weak human and institutional capacities, low and

7

unequally distributed income and scarcity of domestic financial resources. They often suffer from

governance crises, political instability and, in some cases, internal and external conflicts. Their

largely agrarian economies are affected by a vicious cycle of low productivity and low investment.

The category of LDCs consists of 47 countries, out of which 32 are from Sub-Saharan Africa.

Zambia is one of the LDCs located in Sub-Saharan Africa (United Nations Conference on Trade

and Development, 2017)

1.2.2 Zambia’s economic history The setting for all the empirical data studies reported in this thesis is Zambia. Zambia is

a landlocked country covering an area of 752,614 km 2 . According to the 2010 census, the country

had a population of 13.1 million people (Central Statistics Office, 2011). More recent estimates

suggest a population of 16.59 million (World Bank, 2017). It is located in the Southern Africa Sub-

region. Figure 1.1 shows the map of Zambia. Zambia's economy underwent strong growth in the

last fifteen years, with GDP growth between the years 2000 and 2005 averaging 5.8 % per year,

while the 2006 to 2015 period recorded increased growth averaging 6.9% (Ministry of National

Development Planning, 2017). The Zambian economy has been dominated by copper mining

from the time the first commercial copper mine was opened in Luanshya at Roan Antelope in

1928 (Lungu, 2016; Boos and Holm-müller, 2016). Zambia is the largest copper producer in Africa

and the sixth largest in the world. Copper accounted for between 70% and 78% of the country’s

foreign exchange earnings and about 70% of total merchandise exported in the period 2008 to

2016 (Bova, 2012; Trimmer III, 2016; Kragelund, 2017; Bank of Zambia, 2016).

The country also experienced economic growth which led to growth in other sectors

such as manufacturing, construction, trading and transport. During that period, Zambia was

considered to be a middle-income country. The mining sector was owned and controlled by two

private companies, namely the Roan Selection Trust (RST) and the Anglo-American Corporation

(AAC). However, during between 1964 and 1970, the Zambian government led by Kenneth

8

Kaunda became dissatisfied with the level of investment into the mines by both RST and AAC.

The government was also concerned about the distribution of the country’s mineral wealth and

wanted revenue from the mining sector to be used to support other sectors such as education and

health (Kragelund, 2017; Lungu, 2008). Such discontent, coupled with the desire for public

ownership of the means of production in line with the government’s political philosophy at that

time, socialism, resulted in the nationalisation of the mines. Nationalisation was implemented

through economic reforms commonly known as the Mulungushi Reforms of 1968 and the Matero

reforms of 1969 (Chirwa and Odhiambo, 2017). Through these reforms, the Zambian government

sought to develop local industry through increased participation in key economic sectors. The

Zambian government obtained a 51 percent share in mining and other major industries resulting

in the economy being dominated by state-owned-enterprises.

Figure 1.1. Map of Zambia

by state-owned-enterprises. Figure 1.1. Map of Zambia Source:

Source: http://www.un.org/Depts/Cartographic/map/profile/zambia.pdf

9

The nationalisation programme was developed at the time when copper prices were high

and was anchored on a policy agenda that promoted broad social and economic development

(Fessehaie, 2012). Consequently, the government used resources from the mining sector to

support other social sectors such as education and health (Sikamo et al.; 2015, Lungu, 2008).

During the mid-1970s, the mining sector experienced difficulties due to the collapse of copper

prices and the increase in oil prices on the world market (Fraser and Lungu, 2007; Chirwa and

Odhiambo, 2017). However, just as was the case in the pre-nationalisation period, the government

did very little re-investment in mineral exploration and mine development. The mines became

undercapitalised and employed old mining technology using obsolete equipment (Fessehaie, 2012;

Lungu, 2008). Furthermore, the ore deposits became deeper and mineral grades leaner, requiring

modern advanced mining technology to exploit these deposits (Sikamo et al., 2015).

Since Zambia’s economy was heavily dependent on mining (Lungu, 2016; Boos and

Holm-müller, 2016) the resultant fall in production and revenue from the mining sector forced the

government to rely on foreign loans to support health and education facilities. Likewise, other

sectors of the Zambian economy were affected due to their dependency on the performance of

the mining sector. The local industry which was also borne of the Nationalisation programme (that

created state enterprises) was import-dependent and relied heavily on foreign exchange from

copper export earnings. The inability to import inputs both for local industries and mining led to

low capacity utilisation, inefficiency and increased costs in all the state-owned enterprises. The

situation was exuberated by weak management and the conflicting objectives that these

organisations pursued at the time. The enterprises were viewed by government as a way of

employment creation and source of lowly priced consumer goods, while these enterprises also

needed to make a profit for their survival. The country experienced severe economic decline in

the 1980s and 1990s partly due to poor economic management and the decline in international

10

copper prices (Lungu, 2008). The poor performance of the state-owned enterprises led to the

country experiencing severe shortages of consumer goods leading to public discontent. This

culminated in the change of government in 1991.

The new government implemented broad economic reforms that transformed the

economy from a socialist orientation to being market driven. One of the key issues that the reforms

dealt with was the privatisation and commercialisation of state-owned enterprises. This involved

a roll back of the state’s involvement in enterprise management. In the wake of privatisation, state-

owned enterprises collapsed after years of operating under a protectionist economic environment

(Central Intellingecy Agency, 2013; Rakner, 2003; N’gona and Dube, 2012). One of the most

observable effects of the privatisation process were mass job losses in the mining and other sectors

as employment reduced by 66% (Fraser and Lungu, 2007; Gough et al., 2015). The resultant job

losses and reduced employment prospects led to an increase in small firm formation (Kragelund,

2017). It is estimated that small firms accounted for 97% of the total enterprises and 18% of the

total labour force in Zambia in 2007 (Clarke et al., 2010; Ministry of Commerce Trade and Industry,

2009; Phillips and Bhatia-Panthaki, 2007). Because of the change in government policy, many

people who lost employment were pushed into entrepreneurship.

In recent years, Zambia’s economy has experienced strong growth because of the

economic reforms that have been implemented. The country has sustained an impressive record

of macroeconomic stability, achieved single-digit inflation and average Gross Domestic Product

(GDP) growth rates of 6.7% since 2004 (World Bank, 2013c). Zambia has been one of the most

open trade regimes in Africa and is among the top ten African countries on the World Bank’s

“ease of doing business index” (World Bank, 2013a). Zambia has experienced political stability

and no civil conflicts since independence. These factors make Zambia suitable for studying small

firm growth since they reduce external factors that may potentially affect small firm growth and

11

may impair the researchers’ ability to adequately measure firm growth and factors that affect it

(Stenholm et al., 2013; Baum et al., 2001).

1.2.3 Small and Medium Enterprises There is no single universally accepted criterion for defining small firms. Ayyagari et al.

(2007) note that the term SME encompass an array of definitions and measures, differing from

country to country, as well as between the sources reporting SME statistics. The commonly used

criteria classify firms as small, medium or large enterprises based on the number of employees.

Most developed countries set the upper limit of number of employees in small firms between 200-

250 (Organisation for Economic Cooperation and Development, 2004). Most LDCs consider

firms with less than 100 employees as small (Richardson et al., 2004). In Zambia, small firms can

be defined based on the national definition of SMEs, which categorise firms by the total

investments in assets, sales turnover, number of employees and legal status. Firms with up to 100

employees, total investment in assets less than ZMK 2 300,000 (about US$30,000 in 2016) and

annual turnover less than ZMK 800,000 (about US$80,000 in 2016) are categorized as SMEs. A

firm is considered informal if unregistered, the value of its assets is less than ZMK 50,000 (about

US$5,000 in 2016) and employs less than ten people (Ministry of Commerce Trade and Industry,

2004; United Nations Conference on Trade and Development, 2003). Senderovitz (2009) points

out that small firms should be defined based on a deliberate and well-grounded choice, taking the

methodology, purpose and/or content of the study into account. For this study, we adopt Ayyagari

et al.’s (2007) definition of small firms as formal businesses with less than 250 employees.

Formality, in this case, implies being registered and having the necessary permits to operate (Thai

and Turkina, 2014; Webb et al., 2013).

2 ZMK is the abbreviation for the Zambian currency, the Kwacha.

12

1.2.4

Zambian Construction Sector In this thesis, we specifically focused on the construction sector, which has been pivotal

to Zambia’s economic growth in recent years. The sector contributed 21.1% to economic growth

and accounted for 13.0% of the GDP in 2012 (Zambia Development Agency, 2012). Additionally,

all firms in this sector are required by law to register annually with a national regulatory institution

called National Council for Construction (NCC). NCC uses a six-tier grading system based on

number of employees, level of sales and value of assets. The values the criteria take between the

different NCC grades are shown in Appendix 3. In the NCC grading system, grade 1 is the highest

and grade 6 is the lowest. The grades 4, 5 and 6 are reserved for small firms. The NCC grade for

each firm is reviewed annually and a firm with a grade improvement is considered to have grown

in the period under consideration. From the data collected during the annual registration, NCC

has maintained a database of firms in the construction sector in Zambia. The database contains

data such as yearly number of employees, level of sales and value of assets.

1.3 Introduction to the Chapters

This thesis addresses the question of how different factors related to the small firm owner affect

firm growth in Zambia. To answer this question, a number of factors will be considered in the

following

four

empirical

chapters.

They

include:

start-up

motivation,

personal

wealth,

entrepreneurial

orientation,

entrepreneurial

passion,

and

entrepreneurial

alertness.

The

arrangement of the chapters is discussed below.

In chapter 2, we investigate what constitutes small firm growth and how it should be

measured in LDCs. This is necessary because conceptualisation and operationalization of small

firm growth has continued to be problematic, especially for research conducted in non-western

settings such as LDCs. It is also important to recognise that measurement of small firm growth is

affected by context-specific factors such as the level of economic development of the region where

the firm operates. This chapter addresses the research question: What constitutes firm growth in LDCs

13

and how should it be measured? This is vital because before we embark on investigating how individual

level factors affect firm growth; we must first understand what constitutes firm growth and how

to measure it. Therefore, in chapter 2 we develop a scale for measuring small firm growth in

African LDCs.

In chapter 3 we investigate how firm growth translates into economic rewards for the

small firm owner, to further explore the relationship between personal wealth and other firm

growth aspects. This chapter presents an empirical examination of how firm founders extract

wealth from their ventures based on the understanding that increasing personal wealth is one of

the key motives of starting a business, especially in environments with few employment

opportunities such as LDCs. Additionally, having recognised that the amount of wealth extracted

by the founder is related to firm start-up motivation, we further evaluate how start-up motivation

moderates the extraction of wealth from firms by firm founders. The chapter addresses the

research question: What is the relationship between small firm growth and growth in personal wealth of the firm

owner, and how is this relationship affected by the motivation for starting the business?

Chapter 4 focusses on how entrepreneurial practices and behaviour guide the firm’s

growth strategy. The chapter presents an analysis of the effect of entrepreneurial orientation (EO)

on firm growth and the moderating role of the business environment. Although various studies

have investigated the relationship between EO and firm growth, the results have been largely

mixed. The contradictory findings may be due to the context in which a firm operates (Lumpkin

and Dess, 1996). Country-specific business environment conditions have known to affect

entrepreneurial behaviour (Marino et al., 2002; Lee and Peterson, 2000). Nevertheless, the

relationship between EO and firm growth has mostly been studied in western, developed

countries, where the business environment in which the small firms operate is different from that

in LDCs. Therefore, this chapter focusses on how different dimensions of EO affect firm growth

14

in the context of LDCs by seeking answers to the following question: How does entrepreneurial

orientation influence firm growth in LDCs and how does the business environment moderate this relationship?

In chapter 5, the focus shifts to how entrepreneurial passion and alertness of the firm

owners affect firm growth. Entrepreneurial cognition has emerged as one of the major

perspectives for understanding how entrepreneurs as individuals affect the performance of their

firms (Mitchell et al., 2002). This chapter presents an investigation of how firm owner’s

entrepreneurial passion affects firm growth, and the mediating role of entrepreneurial alertness.

As such we address the research question: What is the relationship between entrepreneurial passion and firm

growth and how is the relationship influenced by entrepreneurial alertness?

Chapter 6 presents a summary of the general conclusions from chapter 2, 3, 4 and 5, with

a view to address the overall research question: How do small firm owners affect the growth of their firms

in LDCs? The general theme of this thesis concerns how factors related to entrepreneurial decision

making and behaviour of small firm owners affect the growth of their firms. We first discuss what

constitutes small firm growth in LDCs. In our quest to investigate how firm owners affect the

growth of their firms, we focus on the effects of their decision making and behaviour concerning

the management of their firms. Specifically, we investigate how extraction of wealth from the firm

for personal use is affected by the growth of the small firm. Additionally, we investigate how

entrepreneurial orientation, entrepreneurial alertness and entrepreneurial passion affect firm

growth. It is important to note that these chapters represent stand-alone research papers which

are intended to be self-contained and can be interpreted independently. However, since the setting

for empirical data collection is the same, some overlap and repetition between chapters may occur.

1.4

Methodology The methodological approach used in this thesis is a mixed methodology employing

qualitative and quantitative methods. Qualitative methods were used in the early stages of the field

work to establish the relevance of the variables to be included in the study to the context of LDCs.

15

This process was conducted through expert interviews with various key stakeholders in Zambian

in November 2012. The interviews were mainly aimed at establishing the contextual relevance of

the measures of growth and other constructs. A total of 13 expert interviews were conducted with

officials from small business support institutions and government agencies in charge of SME

policy, local and national chambers of commerce, financial institutions, prominent entrepreneurs

and trade associations and a former Minister of Commerce. Additionally, expert interviews were

also conducted with academicians engaged in small business research.

The expert interviews served as a guide on the relevance of variables included in the

quantitative research in the context of LDCs. This approach of checking which individual variables

were most relevant for this thesis was important because the effect of an entrepreneur’s behaviour

and decision making on firm performance is a complex and context-dependent process. Therefore,

it was important that we considered variables that were relevant to our research setting but are

relatively underexplored in entrepreneurship research.

From the expert interviews, we observed that small firms in Zambia were achieving

varying levels of growth despite initially being similar in size, structure and operating under same

conditions. Data from the interviews also revealed that differences in firm growth levels could be

attributed to actions and decision-making of the entrepreneurs. Additionally, research suggests

that some entrepreneurs become more successful than others due to differences in individual

behaviour and decision making (Baron, 2004a, Shane, 2001). Through the expert interviews, we

identified the individual level factors which were key determinants of firm growth in Zambia. The

findings from the expert interviews and literature review provided a basis for the development of

two questionnaires that were used to collect quantitative data. The identified factors that, among

other factors, influence firm growth are personal wealth extraction, strategic orientation of the

firm owner, and emotions and cognition of the entrepreneur. Accordingly, this thesis focused on

16

how start-up motivation (chapter 3), entrepreneurial orientation (chapter 4), entrepreneurial

passion and alertness (chapter 5) affect firm growth in LDCs.

Quantitative data collection was conducted in two phases. The first phase was preceded

by a pilot study in November 2012. In the pilot phase, we administered the questionnaire to thirty

owner-managers drawn from our sample. The aim was to check the clarity in the wording of the

questions, time taken to answer the questionnaire, and to find the most convenient way of

administering the questionnaire. In March 2013, we embarked on the first wave of data collection.

We collected data on firm growth, firm start-up motivation and entrepreneurial alertness. Firm

growth data was used in chapter 2, data on start-up motivation and entrepreneurial orientation was

used in chapters 3 and 4 respectively. A year later in March 2014, we conducted the second

fieldwork and collected data on firm growth, entrepreneurial passion and entrepreneurial alertness.

The firm growth data collected in the second wave was used in chapters 4 and 5 while the data on

entrepreneurial passion and entrepreneurial alertness was used in chapter 5.

Our sample for the quantitative analysis was drawn for the Zambian construction sector.

All the data used in this thesis were collected from small firms in the Zambian construction sector.

Apart from its economic importance to Zambia, the construction sector was selected because of

the existence of a national regulation that required all firms in the industry to register with the

NCC, which made it easier to select and locate the firms to be included in the study. The

methodology sections in each empirical chapter provide more detailed explanations of the sample,

data collection and methods used to analyse the data. The questionnaires used during data

collection are in Appendix 2 and 3. Now we turn to the chapters representing the four empirical

studies already introduced in section 1.3.

17

Chapter 2 Measuring Small Firm Growth in African Least Developed Countries: Evidence from the Construction Sector in Zambia 3

3 This chapter is based on a paper currently under review (second round) for publication in the African Journal of Management entitled “Measuring Small Firm Growth in African Least Developed Countries: Evidence from the Construction Sector in Zambia” with as authors M. Chabala, L. Paas, E. van Burg, and E. Masurel.

18

Abstract

Conceptualisation and operationalization of small firm growth has continued to be problematic, especially for research conducted in non-traditional settings like African Least Developed Countries (LDCs). This is because the measurement of small firm growth is affected by context-specific factors like level of economic development. In this chapter, we present the development of a unidimensional multiple-item scale for measuring small firm growth in LDCs. Based on an assessment of validity and reliability using growth data from small firms in Zambia, we find that owner-manager's perception of growth in employees, sales, assets, personal wealth and expenditure on purchase of key process inputs is the most suitable way of measuring small firm growth.

19

2.1 Introduction The contribution of entrepreneurship to economic growth in both developed and developing

countries is widely recognised in entrepreneurship research (Acs et al., 2008; Carree and Thurik,

2005; van Stel et al., 2005). It is often argued that economic growth is achievable when

entrepreneurship leads to competitiveness, innovation, and employment creation through small

firm formation and development (Friis et al., 2006; Acs and Storey, 2004). However, small firms

only contribute to economic growth if they survive and grow (Shane, 2009; Wennekers et al., Wong

et al., 2005). Thus, from an economic development perspective, small firm growth is a significant

aspect of the essence of entrepreneurship (Davidsson et al., 2006). This realisation has resulted in

numerous studies on small firm growth in both developing countries (Goedhuys and Sleuwaegen,

2010; Naudé et al., 2010) and developed countries (Davidsson et al., 2010; Leitch et al., 2010;

McKelvie and Wiklund, 2010). Although such studies have provided a substantial body of

knowledge, conceptualisation and operationalization of small firm growth has continued to be

problematic (Davidsson et al., 2013b). Consequently, small firm growth research is often criticised

for

being

“stagnated,

characterised

with

inconclusive

research

results

and

slow

theory

development” (Kiviluoto, 2013 p 569). Leitch et al. (2010) observe that not much is known about

small firm growth and that confusion and misunderstanding surround the phenomenon.

The dominant theoretical concerns in the field relate to incoherence in the way small firm

growth is measured. Davidsson et al. (2007) attribute this to differences in theoretical and

epistemological perspectives and interpretations, operationalizations, empirical contexts, modelling

and analysis approaches, as well as the inherent complexity of the phenomenon itself. Additionally,

there is heterogeneity in form and characteristics of small firms and the environment in which they

operate. All these factors are important elements of small firm growth and their variation affects

its conceptualisation and operationalization (Delmar et al., 2003). As a result, researchers make

21

choices between different small firm growth measurement methods, but there is little guidance on

which choices are most appropriate (Wiklund et al., 2009). As a result, the findings of small firm

growth research are often incomparable, which in turn impedes theory development.

The problem of measuring small firm growth is further compounded when research is

undertaken in contextually different settings such as in Western contexts as well as African Least

Developed Countries (hereafter abbreviated as LDCs). LDCs have different social challenges and

structural characteristics compared to developed countries that have been the focus of most

entrepreneurship research (Naudé et al., 2010; Bruton et al., 2008). Small firm growth is affected by

context-specific factors (Stenholm et al., 2013; Dobbs and Hamilton, 2007; Davidsson and

Klofsten, 2003) such as the national level of economic development (Leitch et al., 2010). Contextual

differences affect key aspects of research design and administration (Kriauciunas et al., 2011;

Natukunda et al., 2016). As such, it is important that conceptualisation and operationalization of

research variables account for the context in which research is conducted. Contextualisation

involves considering the various attributes of the research setting as an integral part of the research

process (Zahra et al., 2014). Therefore, testing and adapting theoretical and epistemological

perspectives and operationalizations used for research in developed countries are important points

to consider when conducting research in African LDCs. Despite this, most entrepreneurship

studies

conducted

in

LDCs

have

adopted

small

firm

growth

conceptualisations

and

operationalizations from developed countries without assessing their suitability. Therefore,

improving the conceptualisation and operationalization of small firm growth from the perspective

of LDCs is still an important aspect in the quest to understand entrepreneurship.

This chapter addresses the research question: What constitutes firm growth in LDCs and how

should it be measured? We present the development of a small firm growth measurement scale that

considers the LDC context. This chapter’s contribution is twofold. First, we assess the suitability

22

of existing small firm growth conceptualisations and operationalizations used in developed

countries for research conducted in LDCs. Second, we evaluate the validity and reliability of small

firm growth measurement scales that specifically consider the LDC context. Although there has

been an increase in research in LDCs with firm growth as a dependent variable, much of it has

been conducted without proper operationalization of firm growth. This chapter provides an

operationalization of small firm growth for future research in LDCs. The chapter also provides

guidance on how to adapt existing operationalizations of small firm growth to different contexts.

The chapter is organized in five main sections. First, in the theory section, we outline the

theoretical foundations of measuring small firm growth and present a review of literature on small

firm growth measurement in entrepreneurship research. Second, the qualitative study section

presents an assessment of the suitability and contextualising of existing small firm growth

measurement methods for LDCs. Third, in the quantitative study section we present an evaluation

of the validity and reliability of small firm measurement scales using growth data from small firms

in an African LDC, Zambia. Fourth, in the final section we discuss the results and draw main

conclusions and address the limitations of the study.

2.2

Theoretical background

2.2.1

Small Firm Growth

Most entrepreneurship scholars converge in their understanding of small firm growth by referring

to the classical work of Edith Penrose (Lockett et al., 2011; Bradley et al., 2011; Leitch et al., 2010;

McKelvie and Wiklund, 2010; Davidsson et al., 2007). Penrose (1959) considered firm growth as

having two meanings; first as “increase in amount” and second, as the “internal process of change”.

The former, which is the most emphasised in entrepreneurship literature (Leitch et al., 2010),

focuses on outcome-based indicators that are used when measuring firm growth. From this

perspective, the common indicators of small firm growth that have been used in research include

23

the change in the number of employees, sales volume, revenues, and profitability (Achtenhagen et

al., 2010; Shepherd and Wiklund, 2009; Dobbs and Hamilton, 2007). Firm growth as an internal

process of change is concerned with how growth is achieved and mainly focuses on growth modes.

Therefore, when studying small firm growth, it is important to understand “what is changing” and

by “how much’. In considering “what is changing”, defining the unit of analysis in which the change

is to be observed is key. The “how much” issue involves deciding on how to measure change.

2.2.1.1 Conceptualisation of small firm growth The heterogeneity in small firms and the variations in growth patterns leads to different

conceptualisations of growth. Researchers decide how to measure growth to suit specific

conditions of their studies. Therefore, measuring small firm growth in entrepreneurship research

is an issue that lacks general agreement on the suitability of the available choices. The key decision

points involve selecting which growth indicator(s) to use, the type of data to collect, and the method

of calculating growth (Delmar et al., 2003). The choices made by researchers lead to different

modelling and analysis procedures, which often produce different results. As a result, the

interpretation and generalizability of research results regarding firm growth has been problematic

(Shepherd and Wiklund, 2009; Dobbs and Hamilton, 2007; Delmar, 2006b).

A number of review articles have analysed and reported the variations in small firm

growth conceptualisation and operationalization in entrepreneurship research. Prominent among

them are studies by Weinzimmer et al. (1998), Shepherd and Wiklund (2009) and Achtenhagen et

al. (2010). These reviews assessed articles with firm growth as a dependent variable that were

published between 1981 and 2008 in leading entrepreneurship and management journals. These

three key papers report a range of growth indicators that included financial and non-financial firm

growth measures. The methods of measuring growth ranged from absolute differences and

regression analyses to subjective perceptions of owner-managers. The review by Shepherd and

24

Wiklund (2009), reviewing 82 articles, found that ‘sales’ was used 61 times, ‘number of employees’

13 times, ‘profit’ 9 times, ‘equity and assets’ 16 times while some other indicators were used 15

times. Achtenhagen et al. (2010) and Weinzimmer et al. (1998) also reported variations in the use of

firm growth indicators and firm growth measurement methods in entrepreneurship research.

We adopted the methodology used by Shepherd and Wiklund (2009) to update the

findings of these studies to include recently published research. We reviewed articles published

between 2007 and 2013 in the current leading entrepreneurship journals, namely Academy of

Management Journal, Entrepreneurship: Theory & Practice, Journal of Business Venturing, Journal of Small

Business Management, Journal of Management Studies, International Small Business Journal, Organization

Science, Small Business Economics, Strategic Management Journal and Entrepreneurship and Regional

Development. Additionally, we analysed how small firm growth has been measured in studies

conducted

in

African

LDCs.

We

found

similar

heterogeneity

in

conceptualisation

and

operationalization of small firm growth. We observed that sales, the number of employees and

assets are still the commonly used indicators of growth. The different methods of measuring small

firm growth still being used include: change measured by absolute difference, relative change,

logarithmic difference and change as perceived by the owner-managers. The results of the analysis

are shown in Table 2.1. Other recent studies that focused on how small firm growth, as a dependent

variable, has been operationalized in recent years show similar results (see Rosenbusch et al., 2011;

Brandstätter, 2011; Unger et al., 2011).

25

Table 2.1. Summary of small firm growth indicators and measuring methods in entrepreneurship research (2007 to 2013) 4

References

Growth indicators

Firm growth

measurement method

Chowdhury (2011)

Revenues

Absolute difference

Bradley, Wiklund, and Shepherd (2011) Brinckmann, Salomo, and Gemuenden

(2011)

Sales

Absolute difference

Employees, sales

Absolute difference

Bruneel, Van de Velde, and Clarysse (2013)

Employees, revenue

Absolute difference

Headd and Kirchhoff (2009) Employees

Relative growth

Stam and Wennberg (2009) Davidsson, Steffens, and Fitzsimmons

(2009)

Employees

Relative growth

Sales

Relative growth

Bogner and Bansal (2007)

Sales

Relative growth

Gilbert, McDougall, and Audretsch (2008)

Sales

Relative growth

Mai and Zheng (2013)

Sales

Relative growth

Watson (2012)

Sales

Relative growth

Delmar and Wiklund (2008) Steffens, Davidsson, and Fitzsimmons

(2009)

Sales, employees

Relative growth

Sales, sales and profit

Relative growth

Tomczyk, Lee, and Winslow (2013)

Sales and employees

Relative growth

Anderson and Eshima (2013)

Sales, employees, market share

Relative growth

Watson (2007)

Total income (sales plus other income)

Relative growth

Goedhuys and Sleuwaegen (2010)

Employees

Logarithm difference

Rauch and Rijsdijk (2013)

Employees

Logarithm difference

Younsuk, Jaeun, and Taejong (2010)

Employees

Logarithm difference

Saridakis, Mole, and Hay (2013)

Employees

Logarithm difference

Chandler, McKelvie, and Davidsson (2009)

Employees, sales

Logarithm difference

Coad et al. (2013)

Sales

Logarithm difference

Coad and Tamvada (2012)

Gross output

Logarithm difference

Davis et al. (2010)

Market share, sales

Perceived change

West and Noel (2009) Overall growth

Perceived change

Oswald, Muse, and Rutherford (2009)

Sales Sales rate, market share, profit rate, overall performance Sales, market share, employees, profitability Sales, profit, transaction volume, income, employees

Perceived change

Tang et al. (2008)

Perceived change

Eddleston et al. (2013)

Perceived change

Gielnik, Zacher, and Frese (2012)

Perceived change

Perceived change and Relative growth Relative growth and absolute difference Relative growth and absolute difference Relative growth and perceived change

Wright et al. (2008)

Employees

Rutherford, Kuratko, and Holt (2008)

Sale, employees

Wiklund, Patzelt, and Shepherd (2009)

Sales, employees

Moreno and Casillas (2008)

Overall growth, sales

4 Table 2.1 continues on the next page

26

References

Growth indicators

Firm growth

measurement method

Wright et al. (2008)

Employees

Perceived change and Relative growth Annual growth rate of retained earnings Geometric growth rate

Hamelin (2013)

Sales, investment rate

Falk (2012)

Employees

Coleman (2007)

Sales

Average yearly change

Pirolo and Presutti (2010)

Sales

Average yearly change

Patel, Fiet, and Sohl (2011)

Employees, number of products

Average yearly change

Wang and Altinay (2012)

Employees, sales

Average yearly change

Zahra and Hayton (2008)

Revenues, sales

Average yearly change

2.2.1.2 Choice of firm growth indicators Small firm growth indicators found in the literature can be broadly categorised as financial

and non-financial. Financial indicators include sales, assets, profit, equity, value added, transaction

volume, stock market performance, and cash flow. Non-financial indicators include the number of

employees, the number of customers, market share, new market entry, and export propensity.

Small firm growth indicators are chosen based on availability of data (Bruton and

Rubanik, 2002), what has been used in prior studies, relationship with the other variables under

study (Davidsson et al., 2007), and relevance for policy makers (Davidsson and Wiklund, 2006).

Sales is the most frequently used indicator followed by number of employees and value of assets

(Bradley et al., 2011; Wiklund et al., 2009). Researchers using sales argue that it is a precursor to

other measures urging that an increase in sales leads to increases in assets and employees and

subsequently rising profits or market share (Davidsson et al., 2007; Delmar et al., 2003). Others

contend that the other two indicators (that is: employees and assets) are more inclined to show

change in inputs than in performance outcomes (Casillas and Moreno, 2010; Davidsson and

Wiklund, 2006). From the foregoing, it can be noted that there is still ambiguity on how to select

growth indicators.

27

2.2.1.3

Method of collecting data There are two prominent methods used for collecting small firm growth data (Coad,

2009). First, data found in secondary sources such as company reports and national databases have

been used. In such cases the data are often termed as “secondary data”. Second, in the absence of

objective data, researchers use self-reported data from owner-managers. Such data is called

“subjective data” because of the bias introduced by relying on respondents’ expectations,

perceptions and knowledge about the firm (Achtenhagen et al., 2010). Subjective data comes in two

forms: the respondent can either report values of growth indicators or his/her perceptions on how

the values have changed over a period of time.

2.2.1.4 Method of measuring small firm growth The method used to measure small firm growth also depends on the type of data collected

by the researchers. In most cases, researchers ask for objective performance data or perceptions of

owner-managers on changes in firm growth indicators. So-called objective performance data can

be obtained from secondary data sources such as firms’ financial reports and tax registration

equipment. However, such data are not always readily available for small firms, especially in LDCs.

In such research settings, researchers usually obtain the relevant data by asking for figures of

performance data from the survey respondents. Such data are then referred to as self-reported

performance data. For both objective performance data and self-reported performance data,

formulas are used to calculate firm growth. Several formulas have been used to calculate small firm

growth in entrepreneurship research. The most commonly used formulas are shown in Table 2.2.

Based on the analysis by Delmar et al. (2003), Shepherd and Wiklund (2009) and Coad (2009), we

identified five formulas that have been prominent in entrepreneurship research. These formulas

include 1) absolute size growth, 2) relative size growth, 3) logarithm difference growth 4) size

weighted growth, and 5) average size growth.

28

In research where data are collected based on perceptions, owner-managers are requested

to estimate growth in the chosen small firm growth indicators on a scale that describes possible

changes. For example, the owner-manager may be asked to rate how sales changed over a period

of time on a five-point Likert scale, with possible responses ranging from “increased very much”

to “decreased very much”. In other research, owner-managers are asked to compare the growth of

their firms with that of competitors or peers in the same industry (Ahmad et al., 2011; Krauss et al.,

2005; Madsen, 2007). Small firm growth measured on Likert scales using the perceptions of the

owner-manager is commonly referred to as perceived growth

Table 2.2. Small firm growth calculation formulas

Firm growth measurement method

Formula

References

Absolute size growth

Relative size growth

"# = "# "# ' (

"# = "# "# ' (

"# '(

100

Delmar, Davidsson, and Gartner (2003), Shepherd and Wiklund (2009), Weinzimmer, Nystrom, and Freeman (1998)

Logarithm difference growth

Coad (2009)

-. = log S -. − log S -.'(

Size weighted growth

"# = ( "# "# '( ) × "# "# '(

Average size growth

"# =

( "# "# ' ( )

1

7

2

( "# + "# '( )

Davidsson et al. (2002) Coad

(2009)

Note: " # = G it = growth in firm i at time t, S it = size in selected indicator at (e.g. sales, number of employees or revenue) time t, Si t-1 = size in selected indicator at time t-1

A number of researchers who use perceived growth justify its appropriateness by the high

correlation that has been shown by prior research between firm growth based on perceptions and

on objective performance data (Casillas and Moreno, 2010; Delmar et al., 2003). Others rationalise

its use by pragmatic reasons such as availability of data and the reluctance among owner-managers

to disclose objective performance data (Anderson and Eshima, 2013).

29

2.2.1.5

Multiple indicator indexes versus independent indicators Some researchers have conceptualised small firm growth as a latent variable (Diambeidou

and Gailly, 2011; Avlonitis and Salavou, 2007) and created multiple indicator indexes to measure

growth. This is based on the understanding that growth can manifest itself in several related but

unique indicators. To illustrate this phenomenon, Davidsson et al. (2007) explain that a firm may

increase turnover through innovative means that do not translate into an increase in assets nor in

employees. Similarly, an increase in the number of employees or assets would not always relate to

the need to meet increased sales. Furthermore, in other types of firms, growth may be moderately

spread across the different indicators (Delmar et al., 2003). Therefore, selecting certain indicators

at the expense of others may lead to failure to capture certain aspects of firm growth properly. In

this situation, combining all indicators into a multiple-indicator index becomes an appropriate

method for capturing growth in all its possible manifestations (Davidsson et al., 2010).

Other researchers select only one small firm growth indicator or a select group of

indicators but analyse the results for each of them independently. Researchers who use the former

justify their choice by referring to the correlations among the different indicators (Casillas and

Moreno, 2010; Delmar et al., 2003). Analysing select groups of indicators is justified from the

perspective that small firm growth is a multi-faceted construct. It is therefore argued that by using

more than one growth indicator, the different aspects of firm growth can be captured (Kiviluoto,

2013; Davidsson et al., 2010).

2.2.1.6 Measuring growth in different contexts There have been efforts to address problems of measuring small firm growth in developed

countries, with some success (Davidsson et al., 2013; Wiklund et al., 2009), but the applicability of

such

methods

in

different

contexts

such

as

African

LDCs

is

under-researched.

Most

entrepreneurship studies focus on developed countries despite the importance of understanding

30

what is going on in LDCs (Naudé, 2010). However, there are apparent differences between small

firms in developed countries and LDCs. Kiggundu (2002) identified such differences in the form,

structure, size, management and purpose of the firm. Small firms in LDCs are usually home based,

family businesses or self-employed individuals; the majority of which experience very low sales and

productivity (Clarke et al., 2010). They often face severe market, institutional and resource

constraints. As a result, many of them operate in the informal sector since these constraints largely

impede their ability to register their firms. The contextual differences between small firms in LDCs

and developed countries can be mainly attributed to differences in national economic development.

The existence of such heterogeneity among small firms and their environments is one of the leading

causes of variability in findings in entrepreneurship research (Soininen et al., 2011; Delmar et al.,

2003). Therefore, it is important that measures of firm growth consider the context in which the

firm is operating. We now turn to the qualitative study we performed to further understand the

conceptualization of firm growth in the context of LDCs.

2.3 Qualitative study

Qualitative interviews provide in-depth descriptions of circumstances, behaviours, attitudes,

thoughts and beliefs of people who have experienced the phenomenon (Patton, 2002). In scale

development and validation, qualitative research is used for generating items to represent the

construct and assess the content validity of the items (DeVellis, 2011; DeVon et al., 2007;

Worthington and Whittaker, 2006; Chandler and Lyon, 2001) Specifically, Hardesty and Bearden

(2004) recommend that research using new, changed, or previously unexamined scale items, should

at a minimum be judged by a panel of experts for content validity. Therefore, we conducted expert

interviews

to

determine

the

suitability

of

small

firm

growth

conceptualisation

and

operationalization from developed countries to LDCs.

31

Experts from small firm support institutions, trade associations, chambers of commerce,

government departments responsible for small business development, prominent entrepreneurs

and academicians were interviewed for their expert opinions on measuring small firm growth in

Zambia (see Table 2.3). During the expert interviews, we focused on assessing whether existing

small firm growth indicators and operationalizations of growth are suitable for LDCs.

Table 2.3. List of experts interviewed

Type of organisation

Name of organisation

Position of expert interviewed

 

Zambia Chamber of Small and Medium Business Associations Zambia Development Agency National Council for Construction Ministry of Commerce, Trade and Industry

Research Officer

Small business support institution

Director of SMEs Executive Director and Registrar Senior Economist Manging Director, Head of SME Banking National Trustee

Past President

Financial Institutions

ZANACO

Zambia Chamber of Commerce and Industry Kitwe District Chamber of Commerce and Industry National Association of Small and Medium Contractors

Trade Associations

Secretary General

Association of Small and Medium Contractors

Chairperson and Secretary General

Entrepreneurs

Association of Building and Civil Engineering Contractors Various

Executive Director

Academics

Copperbelt University

Four owner-managers Two lecturers in School of Business who research and teach entrepreneurship courses

2.3.1 Selection of small firm growth indicators The experts acknowledged all small firm growth indicators from the literature as shown

in Table 2.1. However, not all small firm growth indicators were considered suitable for the

Zambian context. Table 2.4 shows the list of small firm growth indicators and the corresponding

number of experts recommending each indicator as suitable for the Zambian context.

Some of the experts interviewed cautioned against using financial indicators because of

the lack of accurate and reliable data. Small firms in Zambia often operate without credible and

reliable accounting information management systems and are not legally obliged to produce audited

32

financial statements. The experts indicated that owner-managers in Zambia had a tendency of

overstating profits and sales when the information is meant for financial lenders while understating

them if the recipients were tax authorities, policy makers and creditors.

Table 2.4. Number of experts recommending a particular small firm growth indicator

Growth indicator

Responses

Sales Employees Assets *Personal wealth *Expenditure on purchase of key process input Profit Cash flow Credit limit Transaction volume Level of formalisation New market entry Product innovation Value-added Equity Export propensity Financial market performance Market share Number of customers

12

11

10

9

8

5

4

4

3

2

2

2

2

0

0

0

0

0

Note: * represents firm growth indicator added from expert interviews

Accounting ratios, equity and cash flow were considered unrepresentative for the day-to-

day management challenges of the owner-managers. Most owner-managers in Zambia focus on

short-term performance management due to uncertainties in the business environment. One expert

added that the “low levels of financial literacy among the majority of small owner-managers make

the accounting-based measures difficult to use”.

A

number

of

non-financial

indicators

were

deemed

unsuitable

because

of

the

characteristics of small firms in Zambia. They include number of customers, market share, new

market entry, export propensity, and product innovation. For the market-based indicators, it was

noted that most small firms in Zambia had no capacity to monitor and analyse changes in their

33

markets. Export propensity and product innovation were deemed unsuitable because most of the

small business did not export and rarely engaged in activities related to product innovation.

The three commonly used firm growth indicators (number of employees, level of sales

and value of assets) were considered suitable for the Zambian context. Sales were considered to be

the essence of the firm and therefore any observable change would reflect firm growth or decline.

Specifically, for Zambia, it was observed that most owner-managers use their ingenuity to find

work for their firms and are directly involved in sales. The owner-managers also monitor the

performance of the firm through changes in sales. As one respondent indicated:

Small firm owners use their personal connections to source for orders. They know how many orders they get and the value of each order. They often refer to the changes in number of orders and order values when explaining performance.

The experts also observed that change in the number of employees was linked to level of

activity in Zambian small firms. They noted that most small firms used temporary employees to

cater for short-term labour demand fluctuations. They would only hire permanent staff when they

feel the increase in activities is sustainable in the long term. Thus, they recommended the use of

number of employees as an indicator of small firm growth. The value of assets was considered an

important measure of small firm growth. Most owner-managers convert their income into fixed

assets to increase their credit-worthiness. One respondent observed:

It’s common to see new business premises built and cars or other assets bought. They (owner- managers) reinvest in assets to build strong bases for collateral required for future borrowing from financial institutions. Most credit providers have high collateral requirements and investing in fixed assets helps to qualify for credit.

Furthermore, some context-specific growth indicators were identified. They included 1)

change in personal wealth of the owner-manager and 2) expenditure on purchase of key process

input. Using personal wealth as an indicator of small firm growth entails measuring the change in

personal belongings of the entrepreneur resulting from owning the firm. This was justified by the

34

view that most owner-managers do not separate their personal wealth from the firm and vice versa.

As one expert noted:

The premises we use were bought from the money I earned while I was still in employment, why should I charge myself rent? It’s all my money and charging myself doesn’t make sense at all. All these assets are mine! It doesn’t matter if I bought them from my personal income or business income.

In most small firms, assets of the firm and assets of the owner-manager are intertwined.

Measuring small firm growth at firm level has the potential of leaving out a key component of

growth that that may be revealed in changes in personal wealth of the owner-manager.

Expenditure on purchase of key process input was offered as a practical alternative to

using profitability as an indicator of small firm growth. Some experts noted that most small firm

owners monitored expenditure more than profitability. An expert who supported this view

commented:

Most of the entrepreneurs pay a lot more attention to expenditure on inputs than they do with profit. They know, for example, better how much cement they buy than how much they earn from it.

Therefore, they are likely to consider an increase in expenditure on purchase of key

process inputs as an indicator of increased activity.

2.3.2 Operationalization of growth Another key aspect of the expert interviews was to check the suitability of the methods

of operationalizing growth. From extant literature, we found that growth can be measured by

comparing firm performance with that of competitors. For this method, it was observed that most

owner-managers are time-constrained and wary of sharing objective performance data with others.

One expert noted that most small firms in Zambia are still shrouded in secrecy. He explains that

“most small business owners did not know how much their colleagues earn and the range of

activities they undertake”. This reduces their ability to compare their performance relative to

35

competitors. Additionally, they are likely to provide approximate figures and perceived growth

rather than the objective performance data. Two experts who had encountered these challenges

when collecting data also supported this view.

The qualitative study therefore confirmed the three, established small firm growth

indicators (number of employees, level of sales and value of assets) and unveiled two LDC context

specific small firm growth indicators (personal wealth and expenditure on purchase of key process

inputs). These small firm growth indicators have been used to calculate small firm growth in

previous entrepreneurship research. Researchers have used either the formulas shown in Table 2.2

to calculate growth when objective data is available or they have relied on perceptual assessment

of the business owner in the absence of objective data.

2.4 Quantitative study The aim of the quantitative study was to evaluate the dimensionality, validity and reliability

of scales developed from the small firm growth indicators and measurement methods that were

identified from the literature and confirmed through expert interviews. In this section, we present

the methodology and results of the quantitative analysis.

2.4.1

Methodology

2.4.1.1

Sample characteristics Our sample was drawn from firms in the Zambian construction sector. The sample

included firms registered with National Construction Council (NCC) in the years 2009 to 2013 in

grades four to six, which are the appropriate categories for small firms. Additionally, the inclusion

of personal wealth as one of the indicators of small firm growth meant collecting data about the

wealth of the owner-manager. Consequently, we included in our sample only small firms that had

owner-managers (Avlonitis and Salavou, 2007; Lyon et al., 2000).

36

The firms in the sample were drawn from Lusaka and Copperbelt regions. Lusaka is the

capital and largest city of Zambia and is the centre of both commerce and government. Copperbelt

province covers the mineral-rich areas of the Northern parts of Zambia and is considered the

backbone of the economy. The two regions form the commercial and industrial areas in Zambia

and account for 82% of the urban population (Central Statistics Office, 2011) and 56% of the

country’s self-employed population (World Bank, 2013c) and as such provide a fair representation

of small firms in Zambia. Out of the total 235 owner-managers that responded to our

questionnaires 55.3% were from the Copperbelt region while 44.7% came from Lusaka and its

surrounding areas. The average firm age was 10 years. In terms of legal status, 94.5% were limited

companies while the remaining 5.5% described themselves either as sole traders or in a partnership.

2.4.1.2 Data collection Quantitative data were collected using a structured questionnaire between May and

August 2013. Out of the 475 owner-managers that were contacted 235 agreed to participate in the

study. The majority of the owner-managers who participated in the study were male (89.4%); this

can be attributed to the low level of female participation in commercial sectors such as construction

in Sub-Saharan Africa (Bardasi et al., 2011; Liedholm, 2002). Additionally, the median age group of

the owner-managers was 39 to 40 years and 53.2% of all respondents were below the age of 40.

Nearly all respondents (98.3%) had received formal education and 88.9% completed some form of

tertiary education.

All respondents answered all questions on perceptual small firm growth indicators, hence

the 100% item response rates in Table 2.5. However, for the questions that asked for objective

performance data, disclosure rates were lower as shown in Table 2.5. A number of owner-managers

did not provide objective performance data because they were not comfortable with disclosing

such kind of data despite assurances from the side of the researcher; others said they did not have

37

any archival data available. The information in Table 2.5 supports the ideas from literature (Åstebro

and Chen, 2014; Hurst et al., 2013) and the suggestions from the expert interviews that owner-

managers are reluctant to discuss firm growth based on objective performance data.

Table 2.5. Response rates for small firm growth indicators

Growth indicator

Number

Response

Responses

rate

Perceived Growth:

Number of employees

235

100%

Sales

235

100%

Assets

235

100%

Expenditure on purchase of key process input

235

100%

Personal wealth Performance information provided by respondent:

235

100%

Number of employees

217

92.3%

Sales

137

58.3%

Assets

133

56.6%

Expenditure on purchase of key process input

120

51.1%

2.4.2

Measures In entrepreneurship research, growth has been operationalized based on owner-managers’

perceptions of change and as actual change. We used both operationalizations because our research

focus was finding a valid and reliable means of measuring small firm growth in LDCs. We asked

our respondents how they perceived their firm’s growth on each small firm growth indicator. All

perception-based

data

were

collected

using

a

five-point

Likert

scale

that

measured

increase/decrease in employees, sales, assets, expenditure on purchase of key process inputs and

personal wealth. We used questions such as, “how has your sales changed over the last five years

(2009-2013)”, with possible responses on a scale ranging from 1-decreased very much, to 5-

increased very much. We also requested survey-participants to respond on similar Likert scales for

assessing perceived changes in annual figures concerning the number of employees, level of sales,

value of assets and expenditure on purchase of key process inputs, but not for personal wealth

38

because of the difficulties we anticipated in terms of availability and willingness to disclose such

data.

We also used data from the NCC database (see Section 1.2.4) to calculate firm growth for

the purpose of comparing the growth measurement methods with an independent measure of firm

growth used in the Zambian construction sector. Each firm in the sector is given a grade category

that indicates its size and we considered a change in grade as an indicator of change in size.

Therefore, a firm that has increased its size can be considered as having grown over the period

under consideration. Using the NCC grades for each firm for the period 2009 to 2013, we

calculated firm growth by subtracting the firm’s NCC grade in 2013 from the grade in 2009. We

considered firms that had improved their grade as growth firms, while those that remained in the

same grade or moved from a superior to a lower grade as no-growth firms. For example, a firm

that was in grade six in 2009 and in grade four in 2013 was considered to have grown since it had

improved its grade by two (6-4 = 2), while one in grade four in 2009 and grade six in 2013 had

reduced its grade by two (4-6= -2) and was considered a no-growth firm. We created a dummy

variable for allocating each firm in either the “NCC growth” group or the “no NCC growth” group.

We assessed whether our survey-based growth indicators in Table 5 have a positive relationship

with the NCC-growth dummy. For example, if an entrepreneur states that the firms’ number of

employees, sales, assets and the individual’s expenditures and personal wealth have all increased,

this response is deemed more trustworthy when the NCC-dummy also indicates growth for the

firm.

2.4.3

Data analysis Establishing construct validity involves the empirical assessment of the adequacy of a

measurement scale and requires assessing three essential components: dimensionality, reliability

and validity (O'Leary-Kelly and Vokurka, 1998; Venkatraman and Grant, 1986). Construct validity

39

shows the extent to which the measured variables reflect the latent construct they are supposed to

measure (Kline, 2011). We used Exploratory Factor Analysis (EFA) followed by Confirmatory

Factor Analysis (CFA), to establish construct validity of the six growth indicators from Table 2.5.

EFA was used as a preliminary analysis to determine the number of factors and patterns of loadings

that best represented the data (Hair et al., 2010; Worthington and Whittaker, 2006). For this study,

it was important to check the factor structure using EFA since our scales included new indicators

of small firm growth. CFA was then used to assess unidimensionality, reliability and convergent

validity of the scales (Hair et al., 2010; Mullen et al., 2009; Bagozzi et al., 1991).

We also validated our scales by comparing the difference in the mean score of two groups

(one showing growth and the other no growth) that we created from our sample using data from

the NCC database. Diamantopoulos (2005) notes that with the exception of content validity, all

other forms of validity are defined in terms of patterns of relationships with other measures. In

this regard, the concern for validity is to check whether the indicators accurately capture the real-

world phenomena to which they refer (MacKenzie et al., 2011). For this purpose, we compared the

growth measurement scales with the NCC grade growth. We conducted an independent samples

t-test to check whether belonging to the “growth” group or “no-growth” was significantly related

to the scores from the firm growth measurement scales. This relates back to the previously

mentioned example that if an entrepreneur states that the firms’ number of employees, sales, assets

and the individual’s expenditures and personal wealth have all increased, we should find that the

NCC-dummy also indicates growth for the firm.

2.4.4

Results

2.4.4.1

Scale dimensionality

In scale development, it is important to establish dimensionality before testing for reliability and

validity (Schjoedt and Shaver, 2012). A measurement scale that represents a single latent trait is

40

considered to be unidimensional (Hair et al., 2010; Gerbing and Anderson, 1988). Since all firm

growth indicators in Table 2.5 intend to measure small firm growth, we assumed unidimensionality.

Both EFA and CFA were used to assess dimensionality for the firm growth indicators using the

six alternative methods presented in Table 2.2. We used EFA to explore the underlying factor

structure by checking the number of factors produced. To conduct EFA, we used principal

components factor analysis with Varimax rotation using IBM SPSS statistics version 24 software.

The results presented in Table 2.6 show the factor loadings for each firm growth measurement

method. It should be noted that the variable personal wealth was only measured based on owner-

managers perceptions, hence only has factor loadings for this method in the table.

Table 2.6. Result of factor analysis for firm growth measurement scales

 

Standardised factor loadings

 

Variance explained by the first factor (%)

Avg.

Variance

Growth measurement scale

Employees

Sales

Assets

*Purchases

Personal

Explained

 

wealth

AVE

Perceived growth Absolute size growth Relative size growth Logarithm difference growth Size weighted growth Average size growth

0.671

0.883

0.849

0.831

0.724

63.29

0.55

0.763

0.882

0.896

0.964

83.98

0.76

0.375

0.791

0.227

0.866

39.18

0.45

0.688

0.800

0.736

0.851

59.44

0.51

0.839

0.543

0.902

0.740

58.98

0.46

0.762

0.818

0.834

0.832

65.89

0.63

Note: * Expenditure on purchase of key process input

Only one growth measurement scale, relative size growth, did not produce a single factor

structure, with only 39.18% of the variance of the observed growth indicators explained by the first

factor, and was therefore not considered in subsequent analyses. That is, this particular scale

produced two factors, with employees, sales and expenditure on purchase of key process inputs

loading on one factor, and assets loading on the second factor. All the remaining scales produced

single factor solutions explaining at least 58.98% of the total variance. The factor loadings for all

growth indicators in the growth measurement methods that had a single latent variable namely

perceived growth, absolute size growth, logarithm difference growth, size weighted growth and

average size growth, were above 0.5. Hair et al. (2010) recommends this as threshold for achieving

41

significance in factor loadings when the sample size is 200 or more. These results provide initial

support for the unidimensionality of five out of the six measurement scales.

We conducted our CFA using AMOS 22 and Maximum Likelihood Estimation (MLE)

procedures. CFA offers a more stringent assessment of unidimensionality than EFA and is

considered to provide better inferential statistics that allow for assessing construct validity (Gerbing

and Anderson, 1988; Hair et al., 2010; Kline, 2011). The fit indices that are taken into account when

determining model fit were Chi-square ( 2 ) with the associated degrees of freedom (df) and p values,

Tucker Lewis Index (TLI), Comparative Fit Index (CFI) and the Root Mean Square Error of

Approximation (RMSEA). Values larger than 0.97 for the TLI and CFI and less than 0.08 for

RMSEA signify good model fit for samples less than 250 with less than 12 variables (Hair et al.,

2010). Separate measurement models for each growth measurement methods were estimated

independently. Table 2.7 shows the CFA results for the five categories of measures that were

deemed unidimensinal according to the EFA reported in Table 2.6.

Table 2.7. Confirmatory factor analysis results for growth measurement scales

Growth measurement scale

df

x 2

p

TLI

CFI

RMSEA

Perceived growth Absolute size growth Logarithm difference growth Size weighted growth Average size growth

5

9.904

0.078

0.981

0.990

0.065

2

4.146

0.126

0.968

0.994

0.068

2

0.482

0.786

1.072

1.000

0.000

2

13.001

0.002

0.651

0.930

0.153

2

3.586

0.166

0.981

0.990

0.065

The results are shown in Table 2.7 and indicate that unidimensionality was achieved in

four measurement scales namely perceived growth, absolute size growth, logarithm difference

growth and average size growth. The only scale that did not show unidimensionality was size-

weighted growth.

42

2.4.4.2

Scale validity The two important components of scale validity are discriminant validity and convergent

validity. Discriminant validity shows the extent to which conceptually similar concepts are distinct,

while convergent validity measures the extent to which measures of the same concept are correlated

(Hair et al., 2010). Since all our firm growth measurement scales had one latent variable, we only

used convergent validity to establish scale validity. There are two prominent methods for assessing

convergent validity. First, convergent validity is assessed by calculating the average variance in the

indicators that is accounted for by the latent variable (MacKenzie et al., 2011). When each growth

indicator is related to one latent variable, the average variance extracted (AVE) is calculated as the

mean of the sum of the squared standardised factor loading (Fornell and Larcker, 1981). Second,

convergent validity is accessed by considering the size of factor loadings. High and statistically

significant factor loadings show that indicators converge on a common point (Byrne, 2013). Hair

et al. (2010) and Fornell and Larcker (1981) recommend that standardised loading estimates and

AVE values greater than 0.5 show convergent validity.

The results shown earlier in Table 2.6 show that the factor loadings for all growth

indicators in the scales for perceived growth, absolute size growth, logarithm difference growth

and average size growth, had loadings that were above the 0.5 threshold. The values for factor

loadings for sales in the size weighted growth scale were less than 0.5. The AVE value of 0.46 for

size-weighted growth was also below the recommended threshold. These results show evidence of

convergent validity in scales for perceived growth, absolute size difference growth, logarithm

difference growth and average size growth, but not for the size weighted growth. Model fit indices

shown in Table 2.7 also support the validity of these scales.

43

2.4.4.3

Scale reliability Scale reliability indicates the accuracy and precision of a measurement procedure

(DeVellis, 2011; Chandler and Lyon, 2001). Scale reliability was measured using the two commonly

applied indices, Cronbach’s coefficients alpha ( α) and the Composite Reliability, denoted as P(η ).

According to Hair et al. (2010) reliability estimates between 0.6 and 0.7 are acceptable provided

other indicators of the model’s construct validity are good, while estimates above 0.7 suggest good

reliability. The results for Cronbach’s alpha and composite reliability are shown in Table 2.8. The

results indicate that only perceived growth, absolute size growth and average size growth are

reliable according to both indices. The Cronbach’s alpha ( α) and Composite reliability values for

relative growth and size weighted growth were below the cut-off value. Although Cronbach’s alpha

value for logarithm difference growth was above 0.7, its composite reliability value was below 0.6.

Hence, we did not consider the logarithm difference growth to be reliable.

Table 2.8. Reliability assessment results for growth measurement scales

Growth measurement scale

Cronbachs alpha ( α)

Composite Reliability P ( η)

Perceived growth Absolute size growth Relative size growth Logarithm difference growth Size weighted growth Average size growth

0.85

0.62

0.78

0.78

0.01

0.48

0.77

0.59

0.34

0.52

0.76

0.68

2.4.4.4 Comparison with an external measure-NCC growth The data from the NCC database enabled us to compare our scales with the best available

small firm growth measure for the construction sector in Zambia. We compared the three small

firm growth measurement scales (perceived growth, absolute growth, and average size growth) that

were unidimensional, valid and reliable with growth calculated using data from the NCC database.

We used the grade classification from the NCC database for firms in our sample to create the

dichotomous groups of “growth” and “no growth” firms.

44

We conducted an independent samples t-test with the “growth” and “no growth” groups

as the categorical independent variable and the score from the firm measurement scale as the metric

dependent variable. Only the scores measured using the perceived growth scale showed statistically

significant difference for the “growth” group (M=3.782, SD= 0.608) and the “no growth” group

(M= 3.395, S.D 0.903): t(87)= -2.49, p=0.019. Absolute size growth and the average size growth,

for the “growth” and “no growth groups” were not statistically significant. The results of the t-test

are shown in Table 2.9.

Table 2.9. T-test results comparing NCC “growth” and “no growth” Groups

Growth

Grade growth

N

Mean

Std.

t-value

p-value

measurement scale

Group

Deviation

Perceived growth

Growth

43

3.3953

0.90342

No growth

34

3.7824

0.60776

-2.142

0.035

Absolute growth

Employee

Growth

26

793,104

1,841,314

No growth

26

456,083

1,365,770

-1.545