Research consultancy
CHAPTER THREE
RESEARCH METHODOLOGY
Methodology refers to the detailed procedures to be followed to achieve the research objectives. Methodology is important because it presents and describes the approaches and techniques the researcher used to collect data and investigate the research problem. They included the research design, study population, sample size and selection, sampling techniques and procedure, data collection methods, data collection instruments, data quality control (validity and reliability), procedure of data collection, data analysis and measurement of variables.
The study used Cross-Sectional Research Design, they provided an indepth study of a particular situation. The study also used qualitative and quantitative methodologies for data analysis. According to (Kothari, et al, 2005) the research design constitutes the blue print for the collection, measurement and analysis of data. The study employed a Correlation research design. Quantitative method of data collection was employed in order to avoid restriction to only statistical results and this helps the researcher to take into account human behavior. Quantitative approach helped to describe the current conditions and investigate the established relationships between the identified variables (Olver, et al, 2011)
(Amin-Hanjani, et al, 2005) states that population is the aggregate or totality of object or individuals having one or more characteristics in common that are of interest to the researcher and where inferences are to be made. The study was conducted at Send A-Cow head offices in Uganda. The population took categorized into the key and non-key respondents for the study, the key respondents shall entail; top management and the middle management and the none-key respondents shall comprise of; the technical team, implementation team and some beneficiaries. The total population was 100 respondents carefully selected with the guidance from the HR office. The above respondents are selected because they are part of Send A-Cow and are affected by the funding of the organization.
A sample is a selected group of elements from the totality of a population. It is a part of the population that represents it completely (Shukla, 2020) The sample size was selected using Krejcie and Morgan tables (Krejcie et al, 1970) and using the appropriate sampling technique from a population of 100 a sample of 80 was selected and distributed according to the knowledge and experience of the respondents as well as the researcher’s intuition as detailed in the table below.
Table 3.1: Population, sample size sampling techniques
| Category | Target Population | Sample | Sampling Technique |
| Top management | 04 | 03 | Stratified Sampling |
| Middle management | 06 | 05 | Stratified sampling |
| Technical team | 20 | 16 | Stratified sampling |
| Implementation team | 30 | 23 | Stratified sampling |
| Beneficiaries | 40 | 33 | Stratified sampling |
| TOTAL | 100 | 80 |
Source: Created by the researcher with the aid of sample size calculation from Raosoft, Inc (2004)
3.4 Sampling techniques and procedures
The study employed probability sampling technique while selecting the respondents to participate in the study. This was further scaled down to simple random sampling and stratified sampling as explained below.
Probability sampling or random sampling is a sampling technique in which the probability of getting any sample may be calculated (Ragin et al, 2011) These include; Simple random sampling, cluster sampling and stratified sampling. The advantage of probability sampling is its lower cost compared to non-probability sampling.
3.4.1.1 Simple random sampling
Simple random sampling is a technique where each unit of population is given an equal chance of being selected (Shukla, 2020). This technique was employed because according to (Palinkas et al., 2015), it ensures that every member has an equal chance of being recruited into the sample and it is free from subjectivity and free from personal error. Simple random sampling was adopted in selecting the participants from the non-key respondents including the technical team, implementation team.
3.5 Data Collection Methods
Data collection is the process of gathering and measuring information on variables of interest in an established systematic fashion that enables one to answer stated questions, test hypothesis and evaluate outcomes Both primary and secondary data was obtained. In this study, data was collected using the questionnaire, and document review.
A questionnaire is a research instrument consisting of a series of questions and other prompts for the purpose of gathering information from respondents (Kabir, et al 2018)The researcher used the questionnaire because it is practical in collecting large amounts of information. A questionnaire is practical, large amounts of information can be collected, questionnaires data can easily be quantified, it is also a cheap way of collecting data, a large group of respondents is covered within a short time, it also allows in-depth research, to gain firsthand information and more experience over a short period of time. The questionnaire was used to collect data from the non-key respondents.
3.5.2 Documentary Review
Secondary data is obtained through the use of published and unpunished documents (Musinguzi, 2016) Various publication magazines, journal articles, reports, newspapers report and other source of published information on leadership styles and employee turnover was reviewed by the researcher. According to (Somekh & Lewin, 2005) documents, can be helpful in the research design of subsequent primary research and can provide a baseline with which the collected primary data results can be compared to other methods. The issues that was explored was indicated in the study objective or as laid down in the independent and dependent variables of the study.
The key data collection instruments to be used are the questionnaires, interview guide and document review checklist.
A questionnaire is a formulated written set of questions to which respondents record their answers, usually within rather closely defined alternatives (Kothari et al., 2005) He further states that, the questionnaire was used because it is practical in collecting large amounts of information and the data can easily be quantified, it is also a cheap way of collecting data, a large group of respondents is covered within a short time, it also allows in depth research to gain first-hand information and more experience over a short period of time. The researcher chose the questionnaire as an instrument because the study is virtually relational and the tool is an easy method of data collection. The questionnaire consists of closed ended questions purely structured in nature whose variables will be measured on a 5 point Likert scale (5 Strongly Agree, 4 Agree, 3 Not sure, 2 Disagree and 1 Strongly Disagree). The 5 point Likert scale is the most appropriate way to formulate the different questions for measuring different items from different variables. Each variable had a set of questions drawn based on the indicators as illustrated in the conceptual framework in chapter one. The questionnaire was self-administered to the technical team, implementation team and the beneficiaries.
3.6.2 Document Review Check list
A document review checklist is a document that contains a list of documents that the researcher will use during the data collection exercise (Olsen, 2011)The document review list shall be used for purposes of reviewing documentary data. Documentary data was obtained using published and unpublished documents. The researcher reviewed documents such as journal articles, annual school reports, publications on internal control systems and NGO funding. According to (Groves, Fowler, Couper, Lepkowski, & Singer, 2010) documents can be helpful in the research design of subsequent primary research and can provide a baseline with which the collected primary data results can be compared to other methods.
Data control techniques ensure that data collected is valid and reliable; the instruments was first tested to ensure validity and reliability.
The validity of the instrument measures the relevance of the questionnaire items in measuring what they are supposed to measure (Lub, 2015) Validity refers to the truthfulness of findings or the extent to which the instrument is relevant in measuring what it is supposed to measure (Babbie, 2020) The validity of the instruments was tested using the Content Validity Index (CVI) using expert judgment, taking only variables scoring above 0.7 as accepted for social sciences (Amin-Hanjani et al., 2005) The researcher first pre-tested the instruments to some respondents who shall not be part of the study. The researcher also presented the questionnaires to the expert in the field to get an opinion, the data collection tools also tested for completeness before data is analyzed. The researcher used the Content Validity Index calculated as follows;
CVI = Number of relevant items
Total Number of items
Reliability refers to the consistency, stability and repeatability of results (Hair et al (2007). Quantitatively, the reliability of the instruments was established through a pilot test of the questionnaire to ensure consistency and dependability and its ability to tap data that would answer the objectives of the study. The results of the findings were then subjected to a reliability analysis. Reliability was established using the Cronbach’s Alpha Reliability Coefficient test applying the formula below.
The formula of Cronbach’s Alpha Coefficient (K) is
|
|
K ∑ SD2i
K -1 SD2t
α = Alpha coefficient
K = Number of items in the instrument
∑ = Sum
SD²i = Individual item variance
SD²t = Variance of total score
Upon performing the test, if the values attained is 0.7 and above, the items in the instrument was regarded as reliable and the instruments were applied to collect data.
Upon presenting the proposal before a panel to defend it, a cover letter was obtained from Uganda Technology and Management University to permit the researcher to collect data. The covering letter from Uganda Technology and Management University was used during data collection. Participants were given time to respond and after the researcher collected the surveys the next day. The researcher did not offer them any incentives for participating in the research.
According to (Creswell, et at, 2011) data analysis is the process of inspecting, cleansing, transforming and modelling data with the goal of discovering useful information, informing conclusions and supporting decision-making. He further states that statistical analyses are to describe an account for the observed variability in the behavioral data. Data was collected, coded and edited during and after the study to ensure completeness, consistency, accuracy, and removal of errors and omissions. Data analysis therefore involved quantitative analysis.
Data was sorted using the Statistical Package for Social Sciences (SPSS) method. Both Excel and SPSS have a similar feel, with pull-down menus, a host of built-in statistical functions and a spreadsheet format for easy data entry. SPSS has faster and easier basic function access, it has a wider variety of graphs and charts and it is easier to find statistical tests (Musinguzi, 2016) Pearson correlation and regression analysis techniques was used in analyzing data.
Quantitative data got from the questionnaires was computed into frequency counts and percentage. The descriptive statistics shall include use of frequency tables, mean, and standard deviation. The researcher shall adopt Univariate analysis techniques in analyzing data. Univariate analysis is the simplest form of quantitative (statistical) analysis. In addition to frequency distribution, tables, mean, standard deviation and other measures of central tendency was used in data analysis.
The independent variables and the dependent variable measured on a five-point Likert scale 1- disagree, strongly disagree 2-Disagree, 3-Not sure, 4- Agree and 5- Strongly agree. The choice of this measurement is that each point on the scale carries a numerical score, which was used to measure the respondent’s attitudes, and it is the most frequently used summated scale in the study of social attitude. According to Bill (2011), the Likert scale is able to measure perception attitudes, values and behaviors of individuals towards a given phenomenon.
