3.8 Data collection procedure
Data collection will be from both primary and secondary data sources.
Secondary data is the type of data that is readily available but published for reasons other than the research phenomena in context and it’s known to be cost effective as the internet has made it even easier to access this data from different sources (Saunders et al., 2009).
Though according to Saunders et al. (2009) the use of secondary data may be limited by inappropriate variable definitions, outdated and incomplete datasets, this study, I will ensure that relevant and valuable literature of previous research is selected for the literature review, using recent research in the field and peer reviewed articles to increase the validity of this study.
Primary data will be collected through the interviewer- administered questionnaires, in order to specifically address the research objectives and questions. This approach has been chosen because it will enable the researcher to ensure that the respondent is the one targeted which will improve the data reliability and also improve the response rates from the target respondents, even though it’s less cost effective and there is a possibility of interview bias than in self-administered.
Prior to primary data collection, an introduction letter will be obtained from the Uganda management Institute to introduce the researcher to Nakawa division administration for a formal physical access clearance to collect data from the SMEs.
Robson (2002) emphasizes the need to gain acceptance and consent from intended participants in order to gain access to the data that they are able to provide. For this study, this will be achieved by identifying study field interviewers from Nakawa division, who I will trained on translating questions in Luganda the most spoken language by entrepreneurs in the study area, on the approach to seek consent of a respondent and the general ethical considerations including confidentiality and a right to withdrawal from participation.
3.9 Data analysis
According to Saunders et al. (2009) the survey allows collection of quantitative data which can be analyzed quantitatively using descriptive and inferential statistics. Several scholars define data analysis as the process of cleaning, transforming, analyzing, and modelling data gathered in a research.
For this study, the filled pre-coded questionnaires will be periodically reviewed in the field for completeness, consistence and correctness. The collected data will be entered into the Statistical Package for Social Sciences (SPSS) version 20, tested for normality using a histogram and cleaned of outliers before the parametric tests are used for analysis.
Descriptive statistics in form of tables and figures will be used to appropriately summarize responses and present the data set collected for easy understanding and analysis.
Quantitative data will also be analyzed using Multiple Linear Regression Model in order to explain the relationship between one dependent continuous variable and three independent variables
The general form of the Multiple Linear Regression Model is;
Y= ?0+ ?1X1+?2X2+ ?3X3+ e
Y: the dependent variable (Credit Finance Accessibility) expressed as a linear combination of independent variables X1, X2 and X3
?0: The regression constant i.e. Y = ?0 (intercept) when X1, X2, X3….. Xk = 0
?1: Coefficient financial information exchange (independent variable X1)
?2: Coefficient of collateral security requirements (independent variable X2)
?3: Coefficient of human capital (independent variable X3)
e: Error term
Linear regression analysis was used to estimate the coefficients of a linear equation and the Independent variables that best predict the value of the dependent variable. From this model, test of significance at 5% significant level will be conducted on the various variables of this study using coefficient of determination (R2), correlation coefficient (R), F-test and ANOVA table in order to check the significant of the data analyzed.