research methodology

CHAPTER THREE

METHODOLOGY

3.1 Introduction

This chapter presents the research design, study area, population of the study, sample size and techniques, data management and tools for data collection; validity of data collection instruments, research procedure and data management techniques, measurement of variables and ethical considerations.

3.2 Research Design

The study adopted descriptive research design and correlation design A descriptive design is a scientific design that involves the use of observation and descriptions of behaviours of respondents without causing any influence (Babbie, 2007). Descriptive design was chosen because the research is interested in establishing and describing the effect of graduate unemployment on societal wellbeing in Adjumani town council as it exists on the ground without manipulating the conditions or results of the study.

Correlation research design is a research design that involves establishing the degree and direction of association between two or more study variables. The rationale for choosing correlation design is because I am interested in looking at the relationships between the independent and dependent variables.   The study adopted mixed approach of both qualitative and quantitative paradigms as this will enable the two approaches to supplement each other.

3.3 Population

A population is the collection of the elements which has some or other features in common (Singh, 2018). The study population was 240 persons. It included unemployed graduates, their family members and members of the general public represented by opinion leaders in Adjumani town council, Adjumani district. The study targeted the graduates, their families and members of the general public because they are the ones affected directly by unemployment and therefore believed to give better information for this study.

3.4 Determination of Sample Size

Out of the 240 population for the study. 144 sample size was determined by use of Krejice and Morgan’s table of 1970. The table can be found in the appendix.

3.5 Sampling Techniques and Procedures

Source: Nabweggamu 2017

Snowball, purposive sampling and simple random sampling techniques were employed in sampling 144 respondents for the study. Snowball sampling also called chain-referral sampling is a non-probability sampling technique where the existing respondents suggest other respondents needed for to take part in the study (Babbie, 2007). The snowball sampling method was used to select 120 unemployed graduates, because the population for the study was unknown though they are many the unemployed feel bad to openly be referred to as jobless and it was tough to choose and assemble subjects as samples for research. Exponential non-discriminative snowball sampling was used for the study. Identified and recruited the first unemployed graduate who then provided new multiple unemployed graduates. Each new unemployed graduate in turn provided more new unemployed graduates until I obtained enough sample size of unemployed graduates for this study. In addition, I used simple random sampling to select 12 family members of the unemployed graduates, 2 from each of the 6 wards of Adjumani Town council (Adjumani Central, Biyaya, Cesia, Minia, Paridi and Pakondo). I then purposively selected 12 members of the public (opinion leaders) to be interviewed. Being leaders in the society, they were selected for having better picture of what unemployment problems exerts on the society and therefore better placed to share information on the study topic.

3.6 Data Collection Methods

For this study, data was collected from primary and secondary sources. As such the following data collection methods were used; Questionnaire survey method, Interview method of data collection and Documentary Review method. The Questionnaire method was selected for its ability to gather information faster and best for large population while interview method was selected for its ability to gather in-depth information on the study variables free expression of opinions that are vital for the study. Documentary Review method was necessary because it allows the researcher to establish what information is available and those that are missing on the topic of interest.

3.6.1 Questionnaire Method

This involves the use of questionnaire instrument that consists of a number of questions and prompts used to elicit responses from respondents to generate raw data (Abawi, 2017). My major method of data collection in this study was questionnaire method. The researcher used questionnaire survey to elicit responses from120 unemployed graduate individuals to determine the effect of graduate unemployment on them. The choice is made based on its being best for sensitive and more personal information, less costly and can offer consistent and objective views of respondents (Abawi, 2017) and (Kabir, 2016).

3.6.2 Interview Method

Interviews involve presenting verbal stimuli that elicits oral-verbal responses from participants (Abawi, 2017). One-to-one (face to face) interview conducted using open ended questions to collect qualitative data from 12 members of the families of unemployed graduates. This gathered basic personal information on them and 12members of the general public (opinion leaders) were also interviewed as that enable the researcher to obtain first hand behavioural information and those related to their feelings towards the effect of unemployment on the social, economic and psychological wellbeing.

3.6.3 Documentary Review Method

This involves the process of going through existing documents to gather information to support academic work. The researcher was able to access and review district employment record for the last three years, the district strategic plans and district annual plan. The annual crime records for the district was not accessed the national crime report 2017/18 was accessed instead. This enabled me get information that supplement those gathered using the questionnaire survey and the interview methods. According to Mugenda & Mugenda, (2003), using documentary review as a method of data collection helps the researcher to avoid duplication of work done by other researchers.

3.7 Data Collection Instruments

3.7.1 Questionnaire Survey

I used questionnaires with structured and semi-structured questions for the graduates as they are literate and can therefore fill in the questionnaires easily. The researcher used questionnaire survey method due to its advantages of being less expensive, best for sensitive or personal questions, less interviewer biased, permitting anonymity and eliciting more honest answers, as well as ability to offers a consistent and objective view of the respondents (Abawi, 2017).

3.7.2 Interview Guides

The researcher used interview guides comprising of structured and unstructured questions during a face to face interview with family members of unemployed graduates and members of the general public represented by opinion leaders as some of them may not be able to read and write. This tool was chosen due its advantages in the ability to collect viably rich data, spontaneous responses, unrestricted views of respondents and ability to record visually seen responses among others (Kabir, 2016).

3.7.3 Documentary Review Checklist

For secondary data collection, documentary review check list was used to review district documents related to the research especially district employment records, district strategic plan and district annual reports. The use of documentary review as a tool of data collection helps the researcher to avoid duplication of work done by other previous researchers (Abawi, 2017).

3.8. Validity and Reliability of Instruments

3.8.1 Validity

Research tools was first prepared and presented to the supervisors who will check on their correctness. The supervisors’ comments were used to improve the questionnaire by eliminating all errors. In addition, pretesting of questionnaires was also done to identify validity gaps and to make modifications accordingly. The researcher further carried out content validity analysis to ensure validity of the instruments.  Content Validity Index (CVI) was computed using the formula given;

Content Validity Index (CVI) =

The overall (CVI) was obtained for variables (graduate unemployment and for societal wellbeing), both of which were well above 0.70 so instruments were regarded as valid instruments and used for this study (Mugenda & Mugenda, 2003).

3.8.2 Reliability

The questionnaire was pretested using respondents within Adjumani Town Council and Cronbach’s Alpha Reliability Coefficient was computed using SPSS and scores were evaluated. The Alpha Coefficients was found to be above 0.7, the instruments were considered reliable (Sekaran, 2003).

3.9 Research Procedure

The researcher sought for an introduction letter from the Uganda Management Institute administration to help introduce me to Adjumani district authority who granted me permissions and in turn introduced me to the respondents. The researcher then sought informed consent of the respondents, administered the questionnaires to the unemployed graduates and gave them time to fill in the information and after three days collected the completed questionnaires. Similarly, family members and the members of the general public were interviewed on date and time set. This was done to ensure good research practices and ethics. The information gathered was then presented, analyzed and interpreted. The researcher then discussed the findings and draw conclusions based on those findings.

3.10 Data management Techniques

The data collected was collated and presented both quantitatively and qualitatively as discussed in details below:

3.10.1 Quantitative Data

According to Babbie (2010), quantitative data analysis is a way of representing observations numerically so as to describe and explain the reflected phenomena of the observations. The researcher collated the raw quantitative data collected through the use of questionnaire and analyse them using descriptive and inferential statistical tools with the help of SPSS package. The finding was then presented numerically and summarized in percentages and frequencies using tables to ensure easy visual summary while. Such summary of data was then analyzed and interpreted using simple descriptions and explanations and conclusions to the findings deductively drawn.

3.10.2 Qualitative Data Management

Qualitative data involves explaining, understanding and interpreting persons and situations that is being investigated (Babbie, 2010). The researcher analyzed the data obtained from respondents elicited through interviews using content data analysis. The researcher used content data analysis because the study is interested in describing the patterns and commonalities into different themes and sub-themes as per the study objectives.

3.11 Measurement of Variables

I used nominal scale to measurement respondents’ background information which includes a set of characteristics such as age, level of education category of respondents. A number was assigned to categories for easy identification from which respondents selected the true background information that applied to them.

Similarly, graduate unemployment, the independent variable was measured using self-constructed questionnaire with items put on a three-scale coded thus; 1=agreed, 2= Disagreed 3=Not sure, societal wellbeing was measured using self-constructed questionnaire with items put on a five-scale coded; 1=Strongly disagreed, 2= Disagreed 3= Not sure and 4= Agreed while 5= Strongly agreed. In both cases, the respondents indicated their opinion by selecting the appropriate response on the scale.

3.12 Ethical considerations

3.12.1 Approval by Authorities

Kabir (2016) asserts that ethical consideration such as confidentiality, anonymity and consent seeking, among others, be observed during survey exercise. Based on this assertion, the researcher sought approval and permission from the relevant sources right from Uganda Management Institute in form of introductory letter which was presented to the respective authorities in Adjumani District and respondents respectively.

3.12.2 Informed Consent

The researcher briefed respondents on the purpose of the study and categorically declared that the study contents were confidential and used for the purpose of academics and not reprimand. This was emphasized both in the data collection instruments such as the questionnaire and interview guides. It was also orally stressed during face to face interview encounters.

3.12.3 Respect for persons:

Respondents who wanted to remain anonymous were respected with considerable integrity. The researcher treated respondents as persons who were able make decisions and refrained from making choices for them. Respondents were made take part in the research at will. Researcher only guided and protected those not able to make their own decisions. during the research.

3.12.4 Equal Treatment

Respondents were treated equally regardless of their religious, physical, mental social or political sexual orientations by ensuring the use of the same copies of the questionnaire, interview guides and documentary review guides to all respondents (Kabir, 2016).

3.12.5 Falsification

The researcher collected original data and acknowledge all sources of information included in this dissertation and made a formal declaration that the information presented in the dissertation were original and not having been presented by any person for any academic purpose.

3.12.6 Confidentiality

All personal information gathered for the purpose of this academic work was kept confidential and never used to reprimand, defame or malice any respondent in any way. This was ensured by declaring clearly on all the data collection tools and during personal encounters with respondents.

3.12.7 Control of Plagiarism

The researcher adhered to the Uganda Management Institute’s policy of originality check by ensuring that all pieces of work relating to this study was original. Both proposal and dissertation were subjected to Turn it in anti- plagiarism software check. Reports of originality check generated by the software were duly signed and attached as seen in the appendix.