My research employ a Case Study approach and as Yin (2003) aptly puts it, one should consider a Case study design when the focus of the study is to answer how, when it involves real life behaviours that can’t be manipulated, when the contextual condition of the phenomenon study is relevant and also when the boundaries are not clear between this phenomenon and the context. These features greatly correlate with my research since I am seeking to understand how various colonial ills impacted on Zambia. Consequently, I utilised a Single Case Study research design in which the discernible single case was the colonisation of Zambia. The embedded units in the study are represented by my five clusters/themes which are political/governance institutions, prisons ; torture, political participation and rights, labour rights, discrimination and inequalities. I choose to investigate on Zambia primarily because I was familiar with her and it would be easy for me to collect data. Secondly, because like many African countries it seems to have suffered the most irreparable damages and exploits of colonisation as it was mainly used for extraction of resources such as minerals (Acemoglu and Robinson, 2012)
I also designed an analytical framework based on the discourse historical approach. According to Reisigl (2017:45), “discourse historical approach analyses the relationships between the discursive construction of national sameness and difference, which leads to political and social exclusion of specific out-groups.” My choice of this method was motivated by its multidimensional and interdisciplinary incorporation of theory, methodology and empirically based research practices that produce concrete social applications and lead to conclusive findings on historical events in relation to contemporary activities. This was necessary because it allowed me to conceptualize various colonial historic events through a theorized lens of critique (Given, 2008). This helped me to differentiate my study from a mere narration of historical events but rather provided me with a better understanding of the colonial governance era from 1924-1964 and the era between 1964 till now. Its appropriateness for my topics is outlined by (Reisigl, 2017:47) who states that it is mainly used in discourse and discrimination such as racism, ethnicism and nationalism, discourse and politics/policy/polity, nation-building, and in discourse and history such as socialism, fascism, commemoration and history of discourse studies.
5.2 Data Collection, Processing and Sources
Case study extensively makes use of multiple data sources which is a strategy that enhances the credibility of the research (Yin 2003). Thus I used primary sources and secondary sources. However, due to time constraints and difficulties in accessing interviewees at the time, I was only able to conduct the interviews over the phone and through email although am aware of the limitations of using such a method. I used a questionnaire with semi-structured open-ended questions that helped me to provide the interviewees room to fully articulate their responses. Rockman and Aberbach (2002) submit that a semi-structured questionnaire with open ended questions conveys the contextual gradation of the responses and also probes beneath the surface of a response to the reasoning and premises that underlies it. It also helps develop elicit for longer answers. However, in order for my questions not be misinterpreted and to motivate participants to delve deeper into the topic at hand, my questions where stated clearly and allowed them to make open narrations. I avoided the inclusion of additional information in the questions which could narrow the interviewee’s interpretation and constrain their response. This approach eventually helped me to pedantically mitigate the methodological dilemma of details so as to reduce ambiguity and narrowing the interviewee’s interpretation and constraint of their responses.
Bearing in mind that there are many criticism towards data collected from email and telephone interviews mainly due to the interviewer not being able to observe facial expression and body language, make eye contact or hear the voice tone of the interviewee such as visual and nonverbal cues. Denscombe (2003: 51) acknowledges that the quality of response gained through online research is as much the same as the traditional methods. He argues that the same results were reached in several studies that compared, or conducted, both e-mail and face to face interviews. He further argued that, Interviewees via email remained more focused on the interview questions and provides more reflectively dense answers than their face-face counterparts. However, this does not imply that the qualities of face-face interviews are lower, but rather to highlight the benefits of email interview which gives the interviewee time to be more thoughtful and careful in their responses as compared to when being interviewed face to face.
Since I used more than a single method to collect data in my thesis, I had to triangulate the data. This was important for me in assuring validity of the research as well as to capture different dimensions of the same phenomenon. This also helped me to strengthen my perspectives on various aspects of the research. Furthermore, to ensure validity and reliability, I had to conduct verification by systematically checking the data, maintaining focus, and fitting the data in appropriate categories and themes that I had devised. I also constantly monitored and confirmed the theoretical work of analysis and interpretation. By so doing, I was able to identify when to continue, stop or adjust the research process in order to achieve reliability and validity and sustain a certain level of rigor (Bryman, 2012, Creswell, 2014). My idea in this thesis was not to generate an account of historical events but rather understand how these historical events shaped policy and governance in post-colonial period in Zambia.
For the secondary data, I used a lot of academic and non-academic literature from Lund Library and google scholar search. I made the choice of documents that were relevant for my study. The selection was rather strategic and not random. I divided my secondary data into various categories which include: Capabilities approach, development, post-colonial theory literature, colonialism in Africa, human rights and Zambia’s struggle for independence. I chose these categories to cover all the aspects of my study so that my research aim would be well answered. The specific categories of my secondary literature is as follows: Academic literature, policy documents, newspapers articles and reports, journal articles, textbooks, books that interpret, political commentary, biographies, dissertations, historiographical literature, newspaper editorial and book reviews.
5.3 Data analysis
I categorised my research into two major categories which are historical and contemporary events. I coded the data according to those two categories against development and human rights indicators. The categories were formulated based on the prisms of the post-colonial, human rights and capability approach and their various components. I also delineated historical data and using the theories tried to give a theorised interpretation to it using the discourse historical approach. Reisigl & Woodak (2015:96) specify eight steps of a discourse historical approach, some of which I used. The first step was for me to do what they term as activation and consultation of preceding theoretical knowledge. This process involved recollection, reading and discussion of previous research similar to my topics. During this process I also identified my specific themes for my literature.
The second task was to systematically collect data and contextualise the information. This process was mainly guided by the specific data I was looking to collect. Based on the two major categories I had developed I sought data which was appropriate to my research question as well as in accordance with my discourses and the historical discursive events. I aligned and placed the data collected to the appropriate categories mentioned earlier. Thirdly, I had to make data selection and preparation for my analysis. So I started with downsizing the data according to relevant criteria, transcription of the recorded phone responses and sorting the email responses. This was time consuming but relatively an easier step in the process. Based on the literature review and a first skimming of the data, my fourth task was to revert to my research question and reformulate it to specifications of the data I had collected as well as devise appropriately data aligned assumptions for my study.
I was supposed to conduct qualitative pilot analysis which would have allowed testing categories and first assumptions as well as the further specification of assumptions; however this was not necessary for my study as I had already sufficiently reviewed the assumptions during my fourth task. At this stage I went straight into a detailed case study. In case studies doing analysis of evidence remains one of the least developed and most difficult aspects. Marshal and Rossman (2011) and (Yin, 2003) postulate that, the process of bringing order, structure and interpretation to a mass of collected data is messy, ambiguous, time consuming, creative, fascinating and does not proceed in a linear fashion. However, dealing with these aspects I made use of categorical aggregation, a technique which provides linear and organized analyses for the data collected. This technique was first advanced by Robert Stake (1995) used in case analysis where, the researcher seeks assortment of instances from the data while hoping to construe relevant meaning from these instances. Creswell (2007) designates this technique as very informative which draws meanings across multiple instances of the data. And this process was not difficult as I had already done the assortment of data in the earlier steps of the discourse historical approach. Once I had done my case study analysis, I then interpreted the results and contextualised the relevant knowledge in relation to my themes and research aim.