Research consultancy

CHARPTER THREE

METHODOLOGY

3.0     Introduction

Methodology is defined as an approach a researcher uses to investigate a subject (White, B. 2000). It refers to the philosophical basis on which a research is based. Methods on the other hand are particular techniques used to collect data and information. Therefore, the purpose of this chapter is to describe the methodology used in the study.

This chapter presents the research design , study population, sample size and selection, sampling techniques , and procedures , data collection methods and data collection instruments, validity and reliability, the procedure of data collection , data analysis and measurement of variables.

3.1. Research Design

Research design can be defined as the arrangement of conditions for data collection and analysis of data in a manner that aims to combine relevance to research purposed with economy in procedure (Kothari, 2004).

A cross sectional research design was used in the study. The cross sectional research design was adopted because the study intended to conduct an intensive investigation on the variables under study in a particular setup which was the small enterprises in Bushenyi district (Oso & Onen, 2008).

Cross sectional design was added because data was collected at one point in time from a cross section of respondents (Mugenda, 2003). A cross sectional design was used because it explains the cause-effect relationship, and takes place with minimal interference of natural flow of work in the organizations’ environment.  In addition it is relatively cheaper in terms of time and cost as observed by Mugenda (2003) and Sekaran (2004).

In the study, both qualitative and quantitative approaches were used. The two approaches supplement each other and help to reduce the biases in each approach according to Mugenda (2003). Still, according to Amin (2005) the approaches enables the researcher to have a deeper analysis using the inductive and deductive approaches through qualitative and quantitative perspectives which enable the researcher to analyse data from all angles and give a more concrete and realistic description of the findings.

Therefore, the researcher employed the qualitative approach in order to get the opinions of respondents in relation to the variables under study and their understanding on the factors affecting the performance of small enterprises was sought using interviews and documentary reviews.

The quantitative approach was used to obtain the magnitude of the relationship between the variables and this was mainly through questionnaires to the business people. The study was more quantitative than qualitative because most of the data was got from questionnaires. This was because of the number of respondents that responded to questionnaires against that of key informants.

Data was also got through face to face interviews and documentary review. The quantitative method was used to obtain the magnitude of the relationship between variables.

3.2. Study Population

According to Amin (2005), a population is an aggregate or totality of objects or individuals having one or more characteristics in common and which are of interest to the researcher in a given study.

In this study the target population was 200 people composed mainly of the business people and stake holders. It included agro businesses, retail businesses, welding and metal fabrication, management and staff of Pride Microfinance Ishaka branch. The district planner, chief administrative officer, district chairperson, chief finance officer (CFO), the director Ankole Private Sector Foundation, district engineer and the mayor Ishaka Bushenyi municipality were selected as key informants.

3.3. Sample Size and Selection

The researcher agrees with Mugenda and Mugenda (1999) that collecting data from the entire population is not possible due to cost and time constraints. Therefore by studying the sample, the researcher was able to draw conclusions that can be generalized on the population of interest.

A sample size of 127 respondents was selected from the business community in Bushenyi basing on mathematical approach developed by Krejcie and Morgan cited by Amin (2005).  Therefore, with the study population of 200, the estimated sample was 127. The table bellow summarizes the sample size.

Table 1: Sample Size Determination Table

Source:  from Krejcie and Morgan (1970), as cited by Amin E. (2005).

The sample size was chosen because it is difficult to get information from the entire population. The target population was the population to which the researcher generalized the results of the study. The sample was drawn from the target population. Therefore, the sampled results were generalized to the sampled population (Amin, 2005).

3.4. Sampling Techniques

Simple random; in this sampling technique all members of the population have an equal chance of being selected to participate in the study. The technique was selected because it avoids bias and is easy to use (Neumann, 2006). It was therefore used to select respondents in microfinance institutions, agro business, retail businesses and welding and metal fabrication.

Purposive sampling technique was also be used in the study. Purposive sampling according to Neumann (2006) is a non random sample in which the researcher uses a wide range of methods to locate all possible cases of a highly specific and difficult to-reach population. The researcher used purposive technique on the respondents that provided key information that couldn’t be attained by any other method. These respondents included the district officials for key information.

3.5. Data Collection Methods

3.5.1. Documentary Review

Documentary review involved the search and collection of information from the already existing documents and information like books, internet, magazines, newspapers, district literature on business growth and development in Bushenyi and other resource centers.

3.5.1. Questionnaire Survey

This method was used to collect data from respondents especially staff from microfinance institutions, agro business owners, retail shop owners, welding and metal fabrication businesses. Respondents we given a questionnaire containing a number of interrelated questions and items about the factors affecting the performance of small enterprises with questions from which they selected their opinions. This method was preferred because it reduces the cost of data collection; was used on the selected enterprises and individuals as selected in the sample to collect data (Mugenda, 2003).

3.5.3. Interviews

The researcher used face to face interviews with key informants and these were the district officials.  In this method the researcher asked a number of questions to a key informant who responded freely by giving their opinion, perceptions and views on the problem under investigation. This method was used because it provides in-depth data, is more flexible than questionnaires, and yields a higher response rate than questionnaires (Amin; 2005, Mugenda: 2003, Sekaran: 2004).

3.6 Data Collection Instruments

The researcher used both primary and secondary sources. Primary data was collected from the business community and key informants through questionnaires and interviews while secondary sources included records, government publications, websites, the internet and journals. Data instruments which were used included the questionnaires and interview guide.

3.6.1  Documentary Review Checklist

The researcher reviewed documents in form of books, internet, magazines, newspapers, district and government literature that were relevant to the study so as to get a more informed view of the problem.

3.6.2 The Questionnaire

The researcher used a close-ended questionnaire to collect data in the study. This consisted of questions which the respondents answered in writing. The instrument was used because it was cheap compared to other methods. The researcher used self administered questionnaires to collect data. The self administered questionnaires were filled by respondents who were selected to participate in the study. The self administered questionnaires were delivered physically by the researcher and the research assistant to ensure that they are not delayed.

A five Likert scale questionnaire was used. This contained the dimensions of Strongly Agree (SA), Agree (A), Undecided (UD), Disagree (D), and Strongly Disagree (SD). The same instrument was administered to the respondents who would fill their answerers therein. It was divided into five sections and these included A-back ground information, B-microfinance institutions’ practices and the growth of small enterprises, C-government policies and the growth of small enterprises, D-organizational management and the growth of small enterprises and finally, performance of small enterprises.

Questionnaires were used in order to cover a large number of respondents in a relatively short time and there was generation of data because answers were given in their own mood. This however stimulates the respondents because it gave an insight into their feelings and options (Mugenda and Mugenda, 1999).

3.6.3 Interview Guide

An interview guide was designed and used by the researcher to collect qualitative data from key informants in the district. It had key questions on the performance of small enterprises to be asked to key respondents and then it was filled by the researcher while conducting a face to face interview.  This instrument according to Sekaran (2004) enables the researcher to collect rich and detailed data, clarify doubts, and it enables him to establish rapport and motivate respondents to answer questions.  (Yuko & Onen: 2009), argue that this instrument enables the researcher to collect data that can’t be written, and to capture meaning beyond words, and it yields a high response rate.

3.7. Quality control of the instruments (Validity and Reliability)

To ensure the quality control of the instrument, the researcher ensured that the instrument is reliable and valid. Pre-testing of the above instruments was undertaken to measure that the questions were understood by the respondents and that there were no problems with the wording or measurement. This helped to rectify the inadequacies in the instrument before administering them (Sekaran, 2003).

Pretesting was done on 10 persons of a similar sample in the neighboring district of Sheema in Kabwoohe Town Council who never participated in the actual study. The pretest sample was chosen based on Sekaran, (2003) who says that a sample between 1% and 10% is appropriate depending on the sample size. The pre-test subjects were encouraged to make comments and suggestions concerning the instruments, clarity of questions and relevance.

3.7.1 Validity

According to Amin (2005), he defines validity as the ability to produce findings that are in agreement with theoretical or conceptual values; a research instrument is said to be valid if it actually measures what it is supposed to measure with accuracy. The researcher used experts to rate items in the tools. This is supported by Waner (2005), as cited in Barifaijo, Basheka and Oonyu (2010) who assert that Content validity in the study is ensured by using two or more independent experts to judge/rate each item’s relevancy within the research instrument.

Validity was established through a validity test using the Content Validity Index (CVI). This is a tool that is used to gauge the content of items on an empirical measure. It was applied to the formula bellow;

(CVI) =       number of items rated relevant by all judges  

Total number of items in an instrument

That is; CVI =n/N

The instruments for this study were valid to be used since they had a C.V.I of 0.81.

Using the formula;

Judge1. = 29/36= 0.805

Judge2. = 31/36= 0.86

Judge3. = 32/36= 0.88

Therefore the total= 0.805+0.86+0.88 = 2.55/3 = 0.85

In this approach, Subject Matters Experts (SMEs) were asked to indicate whether or not the instrument was valid, and when the CVI is 0.7 and above, the instrument can be considered valid (Amin, 2004). After the comments by SMEs, the result was 0.85. The researcher considered this because Amin proposes that a minimum CVI of 0.7 and above is valid.

3.7.2  Reliability

Reliability addresses whether repeated measurements or assessments provide a consistent result given the same initial circumstances (Last, 2001). Last’s argument  is strengthened by Amin (2005) and Mugenda(2003) who assert that reliability is the degree to which an instrument consistently measures whatever it is measuring and produces similar results if repeatedly used on same respondents under similar procedure .This is achieved by estimating internal consistence and therefore questions measuring the same concept are grouped together and pretested. The following formula below will be used to determine Cronbach’s Alpha Coefficient.

α= k     [1─ ∑ σ²k]

k­1              σ²

Whereby;

∑ σ²k – is the sum of variable of k parts (items) of the test,

σ – Standard deviation of the test,

α- is the Alpha correlation Coefficient

Table 2; Summary of Reliability Statistics

Source: Primary data

If the Alpha is close to one that is above 0.72 then the instrument will be having a high reliability and if the alpha is bellow it will not  be reliable according to Mugenda and Mugenda, (1999). Since Cronbach’s Alpha Coefficient was found to be 0.856 the results were considered valid for the study.

3.8. Procedure of Data Collection

The researcher wrote the proposal and submitted it to the higher degrees department for defense. Upon successful defense before a master’s defense committee, he was given an authority letter which was then presented the relevant people, the small enterprises and government authorities to get permission to collect data from them.

The researcher recruited a research assistant and then trained him for a day on issues about the research that included questionnaires, ethics (confidentiality), creating rapport and editing questionnaires among others. After permission had been granted by the necessary people, the researcher went ahead to create rapport and make appointments and thereafter, proceeded to collect data from the respondents sampled.

Data was then collected, analysed, interpreted and a dissertation report was produced, forwarded to supervisors who made comments. After the comments from supervisors, the researcher submitted a dissertation book to the higher Degrees Department for Viva.

3.9. Measurement of Variables

The term variable refers to an element that take more than one form, and which a researcher measures, controls and manipulates (Neumann, 2006). Nominal scale which indicates the differences among categories like religion; catholic, Jews, Muslim; racial heritage: African Asian, Caucasian, Hispanic etc. The ordinal scale which indicates the difference plus the categories that can be ordered and ranked  like letter grades ;A,B,C,D,E,F opinion measures ; strongly agree, Agree, Disagree, Strongly disagree as given by Neumann (2006)  that form the Likert scale was used. Therefore a five Likert scale  was used to measure responses of respondents in the questionnaire based on continuum of; Strongly Agree(5), Agree(4), Undecided(3), Disagree(2), and Strongly Disagree(1).  This scale enabled respondents to give their views, opinion and attitudes and perceptions about the problem being investigated

3.10. Data Analysis

Sekaran (2003) suggests that analysis is the evaluation of data. It is the process of systematically applying statistical and logical techniques to describe, summarize and compare data

3.10.1 Quantitative Data Analysis;

Quantitative data was collected mainly from close ended type of questions. The researcher ensured that data is coded and categories into useful and relevant data. The data collected from questionnaires was entered in a computer and analysed using the Statistical Package for Social Scientists (SPSS) to find out how microfinance institutions’ practices, government support and organisational management affect the performance of small enterprises. SPSS is the most recommended package for analyzing social science research data (Sekaran, 2003).

Descriptive data was presented in form of summary tables, frequency distribution and percentages and was analysed using Pearson Correlation.

3.10.2 Qualitative data analysis;

The analysis of interview responses from key informants was edited according to the themes developed in the objectives of the study. The information from open ended questions and interview responses was analysed by listing down all respondents’ views under each question or category. The data was harmonized basing on a common view from respondents on the themes and was then interpreted to give meaning.

Therefore, data was first written, coded and analysed using content analysis and thematic analysis. Content analysis is the reading and re-reading of transcripts/data with the aim of establishing similarities, differences so as to develop themes, while in thematic analysis, Barifaijo et’ al (2010:136) argues that data is analysed by theme and this goes on until the researcher is satisfied that no new issues are arising.