Research Plan and Data Collection
Generally, researchers tend to use two main method of analysing data which are quantitative and qualitative analysis (Isadore, 2000). The former, is a positive approach which assumes a hypotheses has been derived from a theory making it deductive in nature. It is often referred to as theory testing and can be summarised in the following order:
Theory
General Hypotheses
Data Collection
Data Analysis
Results
Conclusion
Theory Confirmation/Revision
Qualitative analysis, on the other hand, is inductive in nature. The use of this analysis is not to test a theory, but to develop and explain it and can be summarised in the following order:
Data Collection
Data Analysis
Conclusions
Development of Hypotheses
Leading to Theory Development
Considering the nature of the current study, a qualitative approach is needed as this research is not aimed at analysis an existing theory but at developing one. The main strengths of this method is it helps when studying cases in dept, provides individual data and enable cross comparison and also helps to explain complex data. Nevertheless, the main limitations to this method are data analysis involves a lot of time and results can be influenced by the researcher's personal bias.
The most excellent way of grouping information required is to setup a research plan which means collecting primary and secondary data (Kotler, 1991).
Primary Data
Primary data comprises of initial facts and figures for an explicit function . This type of data can be collected through: observation, interviews, questionnaires and carrying out tests (Kotler, 1991). Primary information will be gathered to achieve the research's needs. Interviews and questionnaires will help gain more in dept data and will facilitate comparison. However, the major constraint is the time and costs involved. There is limited funding and ability to go to China and to 20 Sub-Saharan African countries to collect first hand data. Therefore, this study will use secondary resources to collect valuable and related information.
Secondary Data
Data which already exists which have been gathered for a different reason is called secondary data (Kotler, 1991). The benefits of using secondary information are that gathering them is less expensive, more convenient and less time consuming than gathering primary information.
Secondary data has been collected from various places and various means as discussed below:
The University of Northampton library, UK, was a good source of information offering various business-related academic journals and books. The literature review is based mostly on the articles in these academic journals. Books as well were a source of secondary data helping understanding the different theories and the analysis methods associated to this study.
Secondary data was also gathered from international organisations such as the World Trade Organisation (WTO), Organisation for Economic Cooperation and Development (OECD) and United Nations Conference on Trade and Development (UNCTAD), World Bank, Bank of Mauritius (BOM), Board of Investment (Mauritius), The State Investment Corporation Ltd (Mauritius), The University of Mauritius (UOM), Southern African Development Community (SADC) and The New Partnership for Africa's Development (NEPAD) mostly. Information of Chinese outward foreign direct investment (FDI) was available from China's Ministry of Foreign Trade and Economic Cooperation (MOFTEC).
Although there are many advantages associated with analysing secondary data, disadvantages also occur in this study as discussed below:
Published information is often one or two years old and recent data, i.e. 2009, are only for the first quarter of the year but very rarely from the end of the end. Therefore some figures might be out of date leading to a slight lack for relevance for the study. Also, since there were no studies taking into consideration the South-South Cooperation as a determinant of FDI, this information will have to be calculated.
Collecting information regarding inflows of FDI to Sub-Saharan Africa (SSA) is difficult most studies focus on Africa in general. Data from each individual country is not a available on the internet from the respective governments.
Methodology and Models
Empirical and theoretical works on the various determinants of FDI has already been discussed in the previous chapter and the possible methodology is explain below:
Asiedu (2002) used the ordinary least square analysis (OLS) which was discovered by Carl Friedrich Gauss in 1795. This method assumes that the dependent variables are a linear function of the independent variables and the equation as shown below:
Y = x1+x2…+xn
where x1, x2, xn are constants
The aim is to find the best constants to generate the most accurate results. The model is said to be linear because when the constants are plotted on a graph, it forms a straight line (Clockbackward, 2009). When different types of constants are plotted together, for example, four constants, it forms a plane (Appendix 1). Its advantages are that it is easy to use, provide a visual results which is easy to understand and can be generated on a computer easily. However, it cannot be applied to this study as it cannot be used with too many variables which is why Asiedu (2002) used only 4. Moreover, in real life, nothing is linear which will make the study theoretical and excessive difference in large and small values of constants make the study inaccurate (Clockbackward, 2009).
Fixed and random effects model also known as a mixed model was used by Onyeiwu and Shrestha (2004). Fixed effects are defined as choosing variables constants variables from a criterion (e.g. tax rate is less than 30%) whereas random effects are choosing random variables (e.g. general tax rate). The formula for this method is a vector equation as shown in Appendix 2.
The biggest advantage of this model is that the error vector is related to the individual effects of each factor. This method is no suitable for this research as it have no control over other factors makes it confusing. Since it involves much computation and the level of south-south cooperation cannot be computed easily, it makes it irrelevant.
Hausman specification test was used by Sawkut et al. (2007) in their research because it compares fixed (assumes differences in data) and random (explores differences in error variances) effects of data collected under the null hypothesis, where each effects are uncorrelated in the model (Hausman, 1978). It is used to determine the relationship between an efficient variable and an inefficient one and mostly importantly which of the fixed or random effect model should be used for the study and the equation is shown in Appendix 3.
It is useful to determine which model, either fixed or random to use, but nevertheless, the use of large samples and high degree of freedom in variables which leaves the level of difference undetermined and creates an error coefficient which cannot be valued. Another limitation is that is does not change over time which is the main point of the study considering the time period.
For this research, the empirical work for Cleeve (2008) and Billington (1999) is used to investigate the determinants affecting Chinese FDI in SSA and Mauritius. There will be an improvement of these previous by accounting for South-South Cooperation.
This study will use the cross sectional multiple regression model to evaluate the relationship between FDI from China and the rest of the world to the variables in SSA and Mauritius to the period 2000-2009. It is based on the 10 most successful countries in attracting FDI and the 10 worst successful countries in attracting FDI in the SSA region.
Based on the model of scholars such as Cleeve (2008) and Billington (1999), the linear equation can be formulated as:
FDI = f(X1, X2, X3, ….Xi)
Where: i = 1,2,3,…,n
FDI = FDI inflow from China and the rest of the world to host countries in SSA
Xi = Different variable which influence FDI inflows to host countries in SSA
Billington (1999) used this model to analyse the location of FDI in the UK; Cleeve (2008) used the model to discuss how monetary policies by governments in host countries increases FDI and Ancharaz (2002) used cross-country regression analysis to compare the inflows of FDI in SSA to the rest of the world. To obtain successful results, the location and the different factors affecting the countries in the study needs to be considered and the equation does give a guideline for this.
Chen (1997) suggested that this equation needs to be altered to explain the level of inflows of FDI in a host country. Ordinary least square (OLS) is used to evaluate the power of location factors affecting FDI (Zhao and Zhu, 2000). Variables which are dependent and independent have changed to ordinary logarithm. This will considerably decrease the bend of information when plotted on normal scales. This allows the usage of the regression analysis and also improves integrity in the analysis. The adoption of a log-linear form can be useful to change non-linear bonds between FDI and economical factors affecting FDI into linear ones. This will be the basic functional form linking FDI inflows and economical variables in host countries in SSA (Chen, 1997). The equation will consequently change as follows:
InFDIit = β0 + β1InXit + ε
Where: i = cross-section unit or country
t = time
ε = the error term
The current research is based on the determinants affecting FDI from Chinese companies in SSA as compared to the world and therefore this will be the elementary equation used in this study. The following section will describe the determinants of FDI inflows and set up the independent factors.
Variable Set
The Dependent Variable
The Explanatory Variables
Combine Variables and the Model
(Word Count: 3,955)
Referencing
Bibliography
Krogstrup, S. and Matar, L. (2005). Foreign Direct Investment, Absorptive Capacity and Growth in the Arab World. Graduate Institute of International Studies: Geneva. Working Paper No. 02/2005.
The World Bank Group (2009), Global Investment Promotion Benchmarking 2009: Summary Report. Washington D.C.: The World Bank Group.
United Nations Conference on Trade And Development. (2006) Measuring Restrictions on FDI in Services in Developing Countries and Transition Economies. New York and Geneva: United Nations.
Appendices
