3.0 Introduction
In this chapter, the methods used in carrying out the study are detailed. The chapter outlines the study design, ethical considerations, instrument of data collection, sample size, sampling method, pre-testing procedures, data analysis and quality assurance.
3.1 Study design
This study adopted qualitative case study methodology for a number of reasons. Explicitly, a qualitative approach is warranted when the nature of research questions require exploration (Stake, 1995). Qualitative research questions often begin with how or what, so that the researcher can gain an in-depth understanding of what is going on relative to the topic (Patton, 2002; Seidman, 1998). This study seeks to explore the existence communication models in agriculture, specifically along the honey value chain. It also uses how and what type of questions. Qualitative case study methodology allows the exploration of individuals or organizations, simple through complex interventions, relationships, communities, or programs (Yin, 2003) it supports the deconstruction and the subsequent reconstruction of various phenomena. This study is about Api Trade Africa and the other actors works with along the honey value chain.
This study focuses on how and why questions, the behaviour of Api Trade Africa and other actors cannot be manipulated, furthermore, the contextual conditions in which Api Trade Africa and value chain actors work are relevant to the study.
An exploratory case study was chosen because the case was the communication models in agriculture, specifically the honey value chain but the case could not be considered without the context, Api Trade Africa’s honey production support. It was in these settings that the communication models evolved and were utilized. It is impractical for the study to have a true picture of communication models within the agricultural value chain without considering the context within which the models occurred.
This study explores participants’ perceptions and attitudes towards the various communication models along the honey value chain. Most of the literature that were reviewed used the qualitative method in the data gathering process, including the study by Kigatiira, Mberia, & Wangula, (2018) on the effect of communication channels used between extension officers and farmers on the adoption of Irish potato farming, in Meru County, Kenya.
In determining the case/unit of analysis this study defines a phenomenon as an occurrence in a bounded context. The case is, in effect, the unit of analysis (Miles & Huberman 1994). In the context of this study the case is Api Trade Africa, as a honey support production organization.
This study is alive to the fact that case studies sometimes seek to answer a question that is too wide or a topic that has too many objectives for one study. In order to sidestep this difficulty, a number of authors including Yin (2003) and Stake (1995) proposed the placing of boundaries on a case to prevent this progression from occurring. This study sought to bind the case through; (a) by time and place (Creswell, 2003).
Binding the case ensures that the study remains reasonable in scope. This study covers the period between 2012 and 2018. The selection of 2012 is significant in the sense that the first honey value chain mapping was undertaken in Uganda (Kilimo Trust, 2012).
Secondly, the place of the study is Lira District in Northern Uganda, this is also significant in the sense that Api Trade Africa has worked extensively both directly and indirectly in the district. Northern Uganda is also the highest producer of honey in Uganda (Kilimo Trust, 2012). The boundaries of place and time thus clearly indicate what will and will not be studied in the scope of the study.
This study seeks to explore communication models in agriculture, specifically the honey value chain. There are however intriguing decisions taken by value chain actors that relate to communication that require more attention. This study will thus adopt the exploratory single case study with embedded units. A holistic case study with embedded units enables the study to explore the case while considering the influence of the various communication elements and associated influences on the honey value chain.
3.2 Sample Method and Sample Size
Sampling, as it relates to this study, refers to the selection of individuals, units, and/or settings to be studied. The target population for this exploratory case study comprised honey value chain actors from three counties (Lira Municipality, Agweng and Ogur) of Lira District in Northern Uganda and all the three agricultural production officers involved in the honey value chain. Based on the sampling frame, respondents were selected using purposive sampling. Value chain actors who took part in focus group discussions were selected using homogeneous sampling. Homogenous sampling brings together value chain actors of similar backgrounds and experiences. The study sought to reduces variation, simplify analysis, and facilitate group interviewing (focus group discussions).
This study took into account one primary consideration; the size of the sample being large enough to ensure that all viewpoints have been covered and that the value chain actors have reached saturation point in their responses. This means that the concepts and themes, begin to be redundant. This explains why in each of the three counties a total of four focus group discussions were held and in Lira Municipality five key informant interviews were undertaken.
Since this study adopts the case study methodology, the rule of the thumb based on the approach in this case is to select one case and in this case it is Api Trade Africa Secondly rule of the thumb based on data collection is such that for key informant interviews approximately five respondents would suffice. This study fulfils this rule of the thumb by undertaking five key informant interviews in Lira Municipality. While the focus groups in the three counties had an average of six participants. The recommended rule is to create groups that average five to ten people each. Furthermore, the study based the focus groups on groupings of actors along the honey value chain represented in the research question.
3.3 Data collection instrument and data collection method
The key data collection instruments for this study were one-on-one key informant interviews and focus group schedules. The one-on-one interviews entailed open-ended questions on the honey value chain. Furthermore, these instruments ensured that both the interviewer and interviewee the chance to discuss issues exhaustively. The researcher prepared an outline of open-ended questions to be covered during the FGD. The open-ended discussions provided a forum to a comprehensive exchange on communication along the honey value chain.

In this study, qualitative data collection methods are characterized by the following qualities; open-ended and have less structured protocols i.e., allowing the study to change the data collection strategy by adding, refining, or dropping techniques or informants. This study relies on interactive interviews and it uses triangulation to establish the credibility of their findings (i.e., interviews with key informants to cross check accuracy from actors along the honey value chain.)
3.4 Pre-testing method
This study appreciates the fact that pretesting is essential as a technique for boosting validity in qualitative data collection procedures and the interpretation of findings (Bowden, Fox-Rushby, Nyandieka, & Wanjau, 2002).
Pretesting in the context of this study therefore involved simulating the actual data collection process on a miniature scale to isolate practical glitches with regard to data collection instruments and methodology. The pretest took place in Lira Municipality and involved actors along the value chain. The actors who could speak English understood the interview guide very well. However translation into the local language (Luo) took more time and some parts of the guide had to be rephrased to ease understanding and eliminate errors in cross-cultural language relevance and word ambiguity. The process involved administering the interview to a group of honey value chain actors sharing similar traits to the target study population, replicating how the data collection session was to be undertaken and what type of study materials would be administered including the consent forms and interview guide. The pretest was undertaken methodically and comprised practice for all personnel who were involved in the study’s data collection task. All changes particularly in translation were undertaken based on responses during the pretesting stage.

3.5 Coding ; data analysis
Prior to the commencement of the data collection and the coding process, the study began with a set of priori codes. They were derived from the research questions, problem statement etc. The researcher had some prior knowledge of the subject matter. The second process involved creating codes to be used for the analysis of the case study data and then coding the data. Codes are tags or labels that assign units of meaning to the data and for the quick identification of the segments relating to the research questions and any potential themes (Miles and Huberman, 1994:56).
The purpose of this study’s qualitative analysis is to interpret the data and the resulting themes, to facilitate understanding of the phenomenon being studied. Consequently, this study only used QDA Miner software only to organize the data but not to analyze it.
3.6 Quality assurance
This study worked towards ensuring research quality and rigor. The study used two key strategies to promote rigour and ensure quality assurance. One area that the study paid attention to was the sampling approach and participant selection to enable the research question to be addressed fittingly and diminish the potential of having a biased sample. This was addressed earlier in this chapter. The second area of emphasis for the study was data triangulation, where multiple data sources were used to generate a more in-depth perspective of the case study.
3.7 Ethical considerations
The study sought consent was sought from all actors along the honey value chain before any data was collected from them. Efforts were made to ensure that the diverse actors along the honey value chain understand that the research is principally for academic purpose.
The participants were assured of confidentiality and anonymity. No one was pressured or hoodwinked into participating in the study, this protected both the integrity of the research process and complied with the ethical requirements of a study of this nature.