FIRM PERFORMANCE, CAPITAL STRUCTURE AND CHIEF EXECUTIVE OFFICER TURNOVER 

Get Complete Project Material File(s) Now! »

CHAPTER 4: RESEARCH METHODOLOGY

Introduction

This chapter presents the research philosophy, approach, design and methods used to address the research problem. The two previous chapters reviewed literature on three concepts namely capital structure, performance and change in CEO. In chapter 2, hypotheses emerged. The first hypothesis is about the effect of performance to the amount borrowed by firms and that the performance coefficient can be negative or positive or zero. The second hypothesis is about the effect of leverage on performance propelled by the perceived corporate governance or disciplinary role of debt capital, and that the leverage coefficient can be negative or positive or zero. In chapter 3, three, testable propositions emerged.
The first, (but the third overall) hypothesis is based on the microeconomic theory that predicts that firm performance depends on the managers‘ skills and their effort and therefore, managers whose firms perform poorly are replaced to reverse the decline in performance. The second (but the fourth overall) hypothesis is based on perceived corporate governance or disciplinary role of debt capital and that firms with debt capital in their capital structure are more likely to put pressure on the firm to replace non performing managers. The third (but the fifth overall) hypothesis is the combined and interaction of debt capital and firm performance on the replacement of CEO. The argument is on the possibility that what the shareholders cannot achieve, for example, disciplining managers if the firm is performing poorly, might be strengthened by the presence of debt holders in a firm.
This chapter (chapter 4) linked the preceding chapters to subsequent chapters. It relied on theories in earlier chapters to select appropriate (optimal) research methods required to address research questions, objectives and testing resulting hypothesis as presented in chapters 1, 2 and I output informed the findings and conclusion of this study as presented in chapters 5, 6, 7 and The outputs of this chapter are the research philosophy; research design; population and sample; data and the variables of the study. The other outputs from this chapter were the models’ namely canonical correlation, general linear model (GLM) and generalised estimating equation (GEE).
The rest of this chapter is organized as follows: in section, 4.1 is presentedthe research philosophy; in section, 4.2 are the five hypothesis of the study, as summarised above; section 4.3 are the research design and justification of research design; section 4.4 capture population of the study; section 4.5 is the sample of the study; in section, 4.6 are the data to be collected, validity issues relating to secondary data, data that capture performance, capital structure and change in CEO. Section 4.7 is the variables that capture the three key concepts to be used in testing their relationships. In section 4.8 to 4.9 is presented the method of analysis, and the summary is in section 4.10.

 Research Philosophy

The research philosophies are discussed under the headings’ epistemology, ontology, axiology and doxology and quantitative-qualitative dichotomy (Saunders, Lewis & Thornhill, 2009; Easterby-Smith, Thorpe & Jackson, 2008; Ritchie & Lewis, 2003). A research philosophy is a belief about the way in which data about a phenomenon should be gathered, analysed and used; but science is about transforming things believed (doxology) to things known (epistemology) (Saunders, Lewis &Thornhill, 2009; Easterby-Smith, Thorpe & Jackson, 2008; Hussey & Hussey, 1997; Becker, 1996). In this study, the beliefs are that a relationship exists between capital structure and performance that managers in firms posting poor performance are replaced and that debt holders play a monitoring role, but it is only on the conclusion from this study that the truth of these propositions will surface.
In this study, the core philosophy is that research apart from providing the intellectual resources, contributes to an intellectual rigor and discipline of practical significance. Research is about what is not known; therefore, ‗research philosophy is an over-arching term relating to the development of knowledge and the nature of that knowledge‘ (Saunders, Lewis & Thornhill, 2009).
Researchers can be categorised into two, either positivists (positivism) or non-positivism ornaturalists (interpretivism) (Saunders, Lewis & Thornhill, 2009; Easterby-Smith, Thorpe & Jackson, 2008). Positivists‘ uses quantitative tools and techniques that emphasize measuring and counting. Naturalists prefer the qualitative tools of observation, questioning, and description research. The difference between positivism and interpretivism is extended to their assumptions about what is important to study; what can be known? What are the appropriate research tools and designs? What standards do you apply to judge the quality of the research? In Table 4.1, are the two philosophies presented by Easterby-Smith, Thorpe and Jackson (2008) to guide researchers.
The primary objective of this study is to investigate the relationship between capital structure, performance and replacement of CEO in firms listed on the Nairobi Securities Exchange between the periods 1990-2012. Literature has been used to inform the study, and the study is set to test pre-existing theory relied upon quantitative data, to discover and understand the relationships among the three concepts, performance, capital structure and change of CEO. Therefore, this study adopted a positivist position to address the research problem and research objectives.

Study hypotheses

In order to investigate the relationship between capital structure, performance and replacement of CEO in firms listed on the Nairobi Stock Exchange, the following five hypotheses have been stipulated:
The first hypotheses test the influence of performance on leverage
H₀₁: Firm performance does not have a significant effect on leverage, and alternative
H₁₁: Firm performance has a significant effect on leverage.
The second hypotheses test the influence of leverage on performance
H₀₂: Leverage does not have a significant effect on firm performance; the alternative hypothesis being:
H₁₂: Leverage has a significant effect on firm performance.
The third hypotheses test the effect of performance on change of CEO
H₀₃: Firm performance does not have a significant effect on Change of CEO
H₁₃: Firm performance has significant effect on Change of CEO.
The fourth hypothesis test the effect of leverage on change of CEO
H₀₄: Leverage does not have a significant effect on change of CEO.
H₁₄:  Leverage has significant effect on change of CEO.
The fifth hypotheses test the combined effect of leverage and performance on change of CEO
H₀₅: Leverage and performance has a significant effect on Change of CEO
H₁₅:  Leverage and performance do not have significant effect on Change of CEO.

Research design

Using data in an emerging economy, Kenya, this study established the relationship among three variables namely, capital structure, performance and CEO turnover. This qualified this study to be a correlation (observational) study, which is extended to cause and effect. It is a confirmatory research, to test a-priori hypotheses (Creswell, 2012).
Correlational research was used todiscover or establish the existence of a relationship among variables in this study; that is, between performance, capital structure and change in CEO. The results of correlation research have implications for decision making within businesses; however, the limitation of correlation research is the interpretation of causal relationships. Observational study provides information on what is happening in the real world (Rosenbaum, 2009). In confirmatory studies, hypotheses are usually derived from theories, and then the predictions about the outcomes are made before the measurement phase began. The result of corroborative research is more meaningful in the sense that it is impossible to claim that a certain result is statistically significant or universal unless it is.

Population of the study

The population of this study consists of all firms listed on the Nairobi Stock Exchange (NSE) during the period 1990 to 2012. The 2013 data is excluded because at the time the data was collected, some of the firms delayed releasing their annual reports. As on 31^st December 2012, sixty one (61) firms were listed on the NSE (see appendix 1). Using panel data the study was to employ approximately 1403 (61x23years) CEO years.
NSE listed companies are established firms with ‗elaborate‘ corporate governance procedures, with audited annual report that contained information useful in addressing the objective of this study. The choice of the period 1990 to 2012 is as a result of data availability, taking into account unit of analysis, as a benchmark, in similar studies elsewhere. In Australia, Nielsen and Nielsen, (2013) sampling 146 Swiss listed firms collected data from company annual reports and Web sites on an annual basis for the period 2001–2008; in The Netherlands Glunk and Heijltjes, (2003) studied 60 firms over an 11-year period; in China, though there is no indication of population size, Firth, Fung and Rui (2006) cover a five-year period from 1998 to 2002. In the US based capital structure and performance study, Berger and Bonaccorsi di Patti (2006) employed a sub-sample of 695 banks.

Sampling and sample design

Though all firms listed on the NSE will be included, sample issues arose. The basic sample should enable identification of firms that exhibit the following characteristics over the period of the study: level of performance at both accounting and stock market levels; change in top management and level of borrowing. Such firms must have disclosed the amount of debt and information on top management in their financial statements for the period of the study.
Purposive sampling is used, and out of the sixty one firms listed on the NSE, firms classified as financial institutions are left out, leaving44 firms that translate into 1012 (44×23) possible CEO years, but this depended on availability of data. The next section is on the data required to address research objectives. The data included measures (indicators) of performance and capital structure and change in CEO.

Data collection

The study relied on secondary data (see Appendices 2 and 3 for data collection instruments). The data, specifically market and accounting data required in this study were obtained from the annual reports, copies of which are obtainable through the individual firms, and share price listing found at theNSE and Capital Markets Authority (CMA). Thedata was collectedover the period 1990 to 2012.
Due to lack of depth and thinness of Kenya Capital Market, there were data limitations, that is, issue of sufficient data required to carry a credible study of this level. Therefore, this study employed panel data; that is, instead of a firm being a unit of observation, each firm (or CEO) year during the sample became an observation as in Faleye, Hoitash and Hoitash (2011). As an example, firm X, with 23 years annual variable provides twenty three observation points and not the expected one observation. Panel data is also known as longitudinal or cross sectional time-series data. It is a data set in which the behaviors of organizations are observed across time. The advantage of panel data is that they capture the trend in the study variables in each firm and across firms, and that it is a better way to study the timing of changes in CEO.*
Secondary data have their problems, and it is naive to assume that they are free from errors and flaws (Maxwell, 1996). The main concerns to a researcher relying on secondary data are data validity problems, reliability issues, trustworthiness of data and information, and data source bias. Validity concerns must be addressed because it raises questions on legitimacy of the conclusions that are drawn from data (Trochim, 2006; Maxwell, 1996). Researchers can only defend the use of secondary data when the definitions of a situation by the original data collector match or coincide with that of the theoretical definition of the secondary data user.
Construct validity seeks agreement between concepts expressed by the researcher (constructs) and specific measuring devices or procedures adopted by the researcher. Constructvalidity is identification of data variables that if manipulated will correctly capture the concepts of performance, capital structure and change in CEO. This was attained through literature search (chapters 2 and 3) and adopting standard definitions of performance, capital structure and change in CEO in authoritative studies. This approach took care of content validity concerns.
The comfort in extracting information from annual reports is that they are subjected to an audit by reputable audit firms while the comfort in using market data is that such data is on public domain and is subjected to public scrutiny. An audit lends credibility to information contained in annual reports. However, where returns per share are to be calculated, there will be a need to adjust share prices for dividends and share splits.

DEDICATION 
DECLARATION 
ABSTRACT 
ACKNOWLEDGEMENTS 
CHAPTER ONE:INTRODUCTION 
1.1 BACKGROUND OF THE STUDY
1.2 STATEMENT OF THE PROBLEM
1.3 OBJECTIVES OF THE STUDY
1.4 HYPOTHESES FOR THE STUDY
1.5 VALUE OF THE STUDY
1.6 CONCEPTUAL FRAME WORK
1.7 SCOPE AND DEMARCATION OF THE STUDY
1.8 LIMITATIONS OF THE STUDY
1.9 ETHICAL CONSIDERATIONS
1.10 DIVISION OF CHAPTERS
1.11 SUMMARY OF THE CHAPTER
CHAPTER TWO: CAPITAL STRUCTURE AND FIRM PERFORMANCE 
2.1 INTRODUCTION
2.2 CORPORATE GOVERNANCE AND PERFORMANCE
2.3 CAPITAL STRUCTURE THEORIES
2.4 CAPITAL STRUCTURE AND PERFORMANCE
2.5 MEASURES OF CAPITAL STRUCTURE AND PERFORMANCE
2.6 EMERGING HYPOTHESIS
2.7 CONCLUSION AND SUMMARY OF THE CHAPTER
CHAPTER THREE:FIRM PERFORMANCE, CAPITAL STRUCTURE AND CHIEF EXECUTIVE OFFICER TURNOVER 
3.1 INTRODUCTION
3.2 PERFORMANCE AND MANAGEMENT TURNOVER
3.3 DEBT CAPITAL AND CHANGE IN MANAGEMENT
3.4 PERFORMANCE FOLLOWING CHANGE IN TOP MANAGEMENT
3.5 CHANGE OF CEO AND CHANGE IN CAPITAL STRUCTURE
3.6 EMERGING HYPOTHESIS
3.7 CHAPTER SUMMARY AND CONCLUSION
CHAPTER FOUR:RESEARCH METHODOLOGY 
4.0 INTRODUCTION
4.1 RESEARCH PHILOSOPHY
4.2 STUDY HYPOTHESES
4.3 RESEARCH DESIGN
4.4 POPULATION OF THE STUDY
4.5 SAMPLING AND SAMPLE DESIGN
4.6 DATA COLLECTION
4.7 OPERATIONALISATION AND MEASUREMENTS OF VARIABLES OF THE STUDY
4.8 METHODS OF ANALYSIS.
4.9 CHAPTER SUMMARY
CHAPTER FIVE:CANONICAL CORRELATION BETWEEN CAPITAL STRUCTURE AND FIRM PERFORMANCE 
5.1 INTRODUCTION
5.2 PERFORMANCE AND CAPITAL STRUCTURE INDICATORS
5.3 MEASURES OF CENTRAL TENDENCY – CAPITAL STRUCTURE AND PERFORMANCE VARIABLES
5.4 PEARSON CORRELATION COEFFICIENTS – CAPITAL STRUCTURE MEASURES AND PERFORMANCE MEASURES
5.5 CANONICAL CORRELATION ANALYSIS
5.6 RAW CANONICAL COEFFICIENTS
5.7 Standardised Canonical Coefficients for the Capital Structure Measurements
5.8 STANDARDISED CANONICAL COEFFICIENT FOR PERFORMANCE MEASUREMENTS
5.9 CANONICAL LOADING
5.10 CORRELATIONS BETWEEN THE CAPITAL STRUCTURE MEASUREMENTS AND THE CANONICAL VARIABLES OF THE PERFORMANCE MEASUREMENTS
5.11 CORRELATIONS BETWEEN THE PERFORMANCE MEASUREMENTS AND THE CANONICAL VARIABLES OF THE CAPITAL STRUCTURE MEASUREMENTS
5.12 CANONICAL REDUNDANCY ANALYSIS
5.13 VALIDATION AND DIAGNOSIS OF FINDINGS
5.14 DISCUSSION OF FINDINGS
5.15 THEORETICAL AND PRACTICAL IMPLICATIONS OF THE FINDINGS
5.16 SUMMARY OF THE CHAPTER
CHAPTER SIX: RELATIONSHIP BETWEEN CAPITAL STRUCTURE AND PERFORMANCE USING GENERAL LINEAR MODEL (GLM) 
6.0 INTRODUCTION
6.1 GENERAL LINEAR MODEL
6.2 LEVEL OF MEASUREMENT REQUIREMENT AND SAMPLE SIZE REQUIREMENT
6.3 THE ASSUMPTION OF NORMALITY
6.4 RESEARCH QUESTIONS GUIDING THE ANALYSIS
6.5 INFLUENCE OF PERFORMANCE ON CAPITAL STRUCTURE
6.6 INFLUENCE OF ASSET TURNOVER RATIO (PERFORMANCE) ON TOTAL DEBT TO TOTAL ASSETS RATIO (CAPITAL STRUCTURE)
6.7 INFLUENCE OF CAPITAL STRUCTURE ON PERFORMANCE
6.8 INFLUENCE OF TOTAL DEBT TO TOTAL ASSETS RATIO ON ASSET TURNOVER RATIO
6.9 DISCUSSION OF FINDINGS
6.10 THE THEORETICAL AND PRACTICAL IMPLICATIONS OF THE FINDINGS
6.11 CHAPTER SUMMARY
CHAPTER SEVEN: THE RELATIONSHIP BETWEEN DEBT CAPITAL, PERFORMANCE, AND CHANGE IN THE CHIEF EXECUTIVE OFFICER (CEO) – FINDINGS 
7.1 INTRODUCTION
7.2 SELECTING INDICATORS OF PERFORMANCE AND CAPITAL STRUCTURE
7.3 DATA STRUCTURE AND THE GENERALISED ESTIMATING EQUATIONS (GEE) MODEL
7.4 HYPOTHESIS
7.5 DATA USED IN THE STUDY
7.6 MODEL INFORMATION
7.7 CORRELATED DATA SUMMARY
7.8 VARIABLES OF THE STUDY – CATEGORICAL VARIABLE INFORMATION
7.9 GOODNESS OF FIT
7.10 MODEL RESULTS, INTERPRETATION OF RESULTS AND DISCUSSION
7.11 OWNERSHIP STRUCTURE AND CHANGE OF CEO
7.12 PERFORMANCE AND CHANGE OF CEO
7.13 DEBT CAPITAL AND CHANGE IN CEO
7.14 CHANGES IN CEO, PERFORMANCE AND DEBT CAPITAL
7.15 DISCUSSION AND SUMMARY OF FINDINGS
7.16 THE THEORETICAL AND PRACTICAL IMPLICATION OF THE FINDINGS
7.17 SUMMARY OF THE CHAPTER
CHAPTEREIGHT SUMMARIES, CONCLUSIONS, AND RECOMMENDATIONS 
8.1 INTRODUCTION
8.2 SUMMARY
8.3 CONCLUSION OF EACH OBJECTIVE
8.4 CONTRIBUTION OF THE STUDY
8.5 RECOMMENDATIONS
8.6 LIMITATIONS OF THE STUDY
8.7 AREAS FOR FURTHER RESEARCH
BIBLIOGRAPHY
GET THE COMPLETE PROJECT
The Relationship between Capital Structure, Performance and Replacement of CEO in Firms Listed on the Nairobi Securities Exchange