CHAPTER THREE
METHODOLOGY
3.0Introduction
This chapter presents the methodology which consists of the research design, data types and source, tools of data collection, and data analysis.
3.1 Research design
This study adopted a cross sectional survey design. This design was preferred because it enabled collecting data in a short time (Creswell 2003 and Koul 2005).Quantitative approach was also used because of its flexibility to form multiple scale and indices focused on the same construct (Ahunja 2005).
3.2 Study population
The researcher used secondary data obtained from the national income accounts of Uganda as prepared and compiled by the Ministry of Finance and Economic Development (MOFED), Department of National Accounts, UBOS, UCDA and World Bank Africa database for the period between 1991 and 2010.
3.3 Sample size and selection technique
The study focused on determinants of export volumes of Uganda’s coffee covering 20 years that was from 1991 to 2010. The period was chosen because reforms in the coffee export sub-sector were introduced in 1991. These were expected to have contributed to a sustained increase in coffee exports.
3.4 Data type and sources
Source of data was from secondary sources.The main source of data for this study was the national income accounts of Uganda as prepared and compiled by the Ministry of Finance and Economic Development (MOFED), Department of National Accounts, UBOS, UCDA. In addition World Bank Africa database was used. The data was for the period 1991-2010.
Data on quantity of coffee exports and coffee prices was got from the Uganda coffee development authority (UCDA), Uganda Bureau of statistics, journals, newspapers, text books and while data on interest rates was got from Bank of Uganda and UBOS.Secondary data was sourced because it yielded more accurate information than obtained through primary data, and it was also cheaper.
3.5 Tools of data collection
The data was got by presenting an introduction letter given to me by the head of department Economics and Statistics. This clearly presented my purpose to the different organizations where I waslegible to collect the necessary data for my analysis.
3.6Data Analysis
The time series data was analyzed using Ordinary Least Squares. Unit root tests wasconducted using ADF to find out the existence of unit root. The variables were thentested to determine the presence or absence of cointegration. Error Correction Model was used to capture the short run effects. There after a short-run parsimonious model was estimated after isolating the statistically insignificant variables from the error correction model. This was followed by performing diagnostic tests of serial correlation, stability and heteroskedasticity.The instruments of data analysis will be SPSS, EVIEWS.
