Re-Visiting Ethnographic and Orthodox Research Methodologies: Field Research Experiences from an African Perspective

When undertaking research, there are many things that we take for granted. This concerns research being done either in an African, Asian, Western or any other context. This Chapter shall pay special attention to research carried out in an African context and draws out nuances that pertain to such a setting on the basis that other developing countries in Asia and Latin America might have their distinctive characteristics. Thus, this Chapter is informed by our own experiences as African researchers. A case study based on data from Mashonaland West in Zimbabwe is used to illustrate some of the nuances. Using Western tools of research in an African context has always been a challenge. For instance, English is the main tool of communication in the Western world whereas in Africa, there are different languages or dialects. Researchers have to find some ways to accommodate such specificities as languages or dialects. Women are not allowed to wear pants in certain African settings. Therefore, respondents may refuse to provide information to a woman researcher wearing pants. Yet, doing research while wearing pants, for a woman in a different context may well be acceptable. Casley & Lury (1989) observe that:


Introduction
When undertaking research, there are many things that we take for granted. This concerns research being done either in an African, Asian, Western or any other context. This Chapter shall pay special attention to research carried out in an African context and draws out nuances that pertain to such a setting on the basis that other developing countries in Asia and Latin America might have their distinctive characteristics. Thus, this Chapter is informed by our own experiences as African researchers. A case study based on data from Mashonaland West in Zimbabwe is used to illustrate some of the nuances. Using Western tools of research in an African context has always been a challenge. For instance, English is the main tool of communication in the Western world whereas in Africa, there are different languages or dialects. Researchers have to find some ways to accommodate such specificities as languages or dialects. Women are not allowed to wear pants in certain African settings. Therefore, respondents may refuse to provide information to a woman researcher wearing pants. Yet, doing research while wearing pants, for a woman in a different context may well be acceptable. Casley & Lury (1989) observe that: The special difficulties of conducting surveys in developing countries derive from their socio-economic structure. Agriculture is still the main occupation of most of the population and it presents particular problems. Climatic and soil conditions may vary within one country. Some areas may be inhabited by nomadic pastoralists and others by small farmers. Thus there will often be wide regional disparities particularly since the differences in physical conditions will usually be accompanied by cultural differences (Casley et al., 1989).
There are no traditional barriers when negotiating entry for research in Western countries. In most cases the major barriers are the Ethical Committees at local level and satisfaction of the legal requirements such as the Data Protection Act and the Human Rights Act among others in the case of the UK. In Africa, a researcher has to negotiate with local Chiefs and some very limited extent with the government through ministries or departments of health if the research involves human subjects. This explains why most drug tests are done in Africa and this suggests some exploitable and slipshod laxity on the part of African governments on that score. At Universities, researchers have to obtain permission from the relevant Ethical Committees of that University who may approve or not approve the projects based on its relevance and quality and the treatment of human subjects in a humane way and guarantees of the observation of its ethical codes and standards. Furthermore, the poor state of infrastructure in rural areas of most African countries makes access to respondents very difficult. However, the major challenges faced by researchers when doing research in an African context relate primarily to data collection and sampling. This Chapter will attempt to unpack those challenges and dilemmas when doing research in such a context using essentially an interpretivist paradigm. The key components of the paradigm are 'subjective perceptions and understanding; which arise from experience; objective actions or behavior; and context ' Ulin et al., (2002:22). The Chapter will also posit a few alternative methods with respect to data collection.

Questionnaire dilemmas
Western research methodologies do not take into account African development stages/dilemmas. Africa is a continent clamouring for education, technology, research capacity and skills. It is on these pre-requisites of development, among others, that she is lagging behind. In the rural areas of Africa, illiteracy is still rampant. The challenges experienced when using a questionnaire in an African context may relate to misinterpretation of the questions due to low levels of literacy. This implies that content errors will be committed as a result of the misinterpretations. Questions may also not be answered leading to non-responses. Whole questionnaires may also not be completed. Furthermore, one has to take into account the challenges of language. The multiplicity of languages in one country may present a challenge to researchers. South Africa has eleven official languages. According to the Ethnologue -Languages of the World, the number of individual languages listed for Zimbabwe is 20. Of those, 19 are living languages and one (1) is a second language without mother-tongue speakers (Lewis, 2009). However, Zimbabwe has two dominant languages, namely Shona and Ndebele with a multiplicity of dialects. This means that the research teams inevitably have to be multi-lingual in composition. Language use in the design of the questionnaire itself is also a problem. Care has to be taken with regard to wording and general phraseology. Different words mean different things to different people. For instance, the "you" would be very disrespectful in a research context which in Shona, a dominant indigenous language in Zimbabwe, would be, "iwe". A researcher cannot use such a word in research questions as that may offend the respondent if s/he is a household leader who would prefer to be called "imi", its "you" but used in its plural form as if addressing many people.
When the questionnaire is researcher administered, there is always the danger of the researcher bringing in his/her biases into the research. Most qualitative research is pragmatic from the point of view of the researcher and subjective from the point of view of the subjects. However, the researcher must strive to desist from leading the respondents and bringing in his/her bias into the research. Seasoned researchers have the capacity to do so which comes through (self-) training and experience. Structured questionnaires tend to channel people in a particular direction, directive approach, which are researcher-defined, and usurp discretion from the respondents who will withhold his/her "views', if not covered in the questionnaire, due to these 'structured responses' created by the researcher. It must be admitted that structured questionnaires are easy to analyse as compared to the unstructured ones. The trade off would be between the unquestionable ease of analysis provided by structured questionnaires or depth and variety of responses provided in unstructured questionnaires. These decisions rest squarely with the researcher.
Because of poor literacy levels in a rural African context, there is need to dedicate more time for holding the interviews or for questionnaire administration. Unstructured questionnaires and interviews give more discretion to the respondent to delve into issues based on their own volition, non-directive approach. This has often been found to be the case. Alternatively, respondents because of being "over-researched" tend to provide the researcher with the answers that the researcher expects in order to get rid of him/her as s o o n a s i s p o s s i b l e . T o g a i n t h e p e o p l e ' s confidence the research has to ensure the confidentiality and anonymity of respondents and the researcher must remain genuine/authentic and honest and also perhaps allow respondents to choose their own pseudonyms for the purposes of the study. Creating research partnerships, between the respondents, who purvey the information, and the researcher/learner/co-interpreter, carries with it serious ethical obligations anchored on mutual trust and understanding of common interests (Ulin et al., 2002). Hammersley & Atkinson (cited in Silverman, 2001) posit that respondents are more concerned with the researcher as a person than the research itself -whether 'he or she can be trusted, what he or she might be able to offer as an acquaintance or a friend, and perhaps also how easily he or she could be manipulated or exploited'. The advantage of an 'outsider' is that respondents may be more inclined to discuss issues of a personal nature such as love and lust with them than with an 'insider' to avoid unnecessary mortification and humiliation. The bottom-line is that the questionnaire should better be administered by the researcher in cases where illiteracy is commonplace.

Interviewing conundrum
Interviewing is the process of directing a conversation to collect information (Angrosino, 2007). In an African setting we observe that interviews have their own advantages and disadvantages. One of the disadvantages is that the interviewer may push for his agendathat is, introduce interviewer bias as mentioned earlier. The advantage is that interviews cater for all people irrespective of their background, education, gender and other considerations. Face to face interviews are ideal for matters of a personal nature such as sexuality which should be understood based on personal experience as well as what people and society say about it -personal experience and what people say inform behavior. Telephonic interviews are hampered by the paucity of technology because in the rural areas, homes do not have telephones in many an African context.

Focus group enigma
Some topics are not ideal for Focus Group Discussions (FGDs). People cannot freely air their views. Some people may not feel free to answer questions in front of others and thus may not fully participate in the discussions in spite of the noble objectives the discussions/interviews may have. People may not be comfortable either in the company of others or the researcher herself. Some do not speak out because of lack of self confidence or plain respectfulness, timidity, and even reticence. Those who have the confidence, enthusiasm and buoyancy, dominate discussions with potential to hijack it. Some moderate their answers in the presence of peers. Group interviews tend to exert influence on respondents and in turn influence the outcomes of the research. Some tend to toe the line in the presence of others. Mixed gender groups may be considered appropriate in some contexts but not mixed age groups because of the sensitivity of certain customs. (By way of caveat, even mixed gender groups should be carefully considered because in some settings, African men prefer to have their early morning chats in the kraal with the exclusion of women. Their inclusion relies on the topic to be discussed). Even then, focus group discussions in mixed groups are characterized by male domination. Moreover, for cultural reasons, some people may not want to be isolated and be put into a group of people whose values they do not share. For example, Muslim women in Northern Nigeria would not want to be put in the same focus group with Christian women from the South and by extension, Muslim men may not want to be grouped with Muslim women. These create problems. Thus FGDs are important and relevant for matters of mutual interest such as community, social events but judiciously constructed.

Participant observation riddle
According to Angrosino (2007), observation is the act of perceiving the activities and interrelationships of people in the field setting. Participant observation is unique in that it combines the researcher's participation in the lives of the people under study while also maintaining a professional distance (Fetterman, 1998). In an African setting, those who are being observed may not feel free to be in the company of a stranger. To go about this, it may be meritable to employ researchers who are locals because they can literally 'connect' with their people. This could be particularly true for studies that seek to understand cultural aspects of a people. The disadvantage of using the services of local people is that the obvious things to them may be unique to a stranger and thus interesting to research. This may result in rich grassroot level data not being captured by using their services. Casley et al., (1989) claim that 'insiders' may 'over-identify with the subjects' such that their lines of inquiry are restricted. People tend not to trust strangers in their midst leading to deception of researchers. Furthermore, people usually want to be compensated for their time. Some of this could be attributable to economic disadvantage and some of it as due payment/recognition for effort.

Cultural divergence
There are differences in how issues can be conceived in an African as opposed to a Western perspective. The differences could be cultural/traditional, literacy and moral differences. A culture (Western) that allows both males and females to wear pants will not have a problem with a woman researcher to wear pants when doing research. As earlier mentioned, a culture that does not allow women wearing pants in an African context may not allow enumeration to be done by a woman wearing them. The appropriate dress code has to be observed. Makoni (n.d) observes that euthanasia, homosexuality and sex change are embraced in the Netherlands but are considered taboo in Zimbabwe. This is a reflection of the legal and moral schism between the two countries.
Elderly people do not want to be "cross-examined" by the young ones on moral and cultural grounds, for example, on issues of family, governance and sex-related matters. They do not take to it lightly. Questions related to those matters should not be assigned to young researchers lest they be rebuffed and snubbed. The elders are the custodians of cultural heritage and indigenous knowledge of their communities which they pass on from generation to generation. Thus, secrecy is commonplace regarding ages-old tradition and taboos such that it becomes difficult to obtain credible information on such traditions and taboos. There are topics which cannot be discussed publicly or shared in public because they are thought to be sacred and sacrosanct. For example the initiation ceremonies among the Venda people in South Africa (most ethnic groups in Africa actually try by all means possible to keep their ceremonies sacred and guarded). Participants to those ceremonies do not easily share that information. This is different from Western respondents. One would find that some teenager girls do video tape their debut sex encounter to share with their friends. Issues related to sex and sexuality are neither discussed between fathers and sons, nor can that conversation transpire between mothers and sons (and fathers and daughters) in an African context. It may be slightly different in urban areas where western culture has been adopted. However, this shows that the level of openness differs in each of these two contexts. Cultural prohibitions on divulging issues related to initiation schools, mean that accessibility to facts becomes problematic. Even when respondents agree to be interviewed, the reliability of the information provided may still remain questionable given the 'sacredness' of the practice which is guarded and preserved jealously. Information will not be provided to a non-initiate because of the thorough screening process which involves elaborate test questions. It is unethical to use hidden cameras. Therefore there are those things that are considered to be sacred and therefore are held in secrecy. Those things become very difficult to observe overtly. Covert access, which is access secured without the respondent's knowledge, raises ethical issues and may well endanger the researcher's life if the researcher is found out and his/her disguise is publicly unmasked. Because Africans secretly guard these practices, sexuality among blacks is portrayed as uncultivated, unfettered and insatiable against a western self-portrait of 'restraint, purity, calculation and rationality' (Marais, 2005:15). Such ostentatious and conceited inferences are unwarranted and misplaced to the extent that such generalizations inspire wrong conclusions.
Diversity is not always embraced in certain settings as people tend to look inward to their own cultural and traditional practices to the exclusion of other perspectives. In that sense inclusivity is wantonly and shamelessly disregarded. Furthermore, many things are not properly documented in an African context, this makes researching them very difficult in cases of secondary data search.

Seasonality
Most African contexts are affected by seasons as in other contexts. There are times when people are too busy that they can hardly have time to take questions for example during rainy season and harvesting periods. Researchers have to accommodate these periods. This is particularly true when the researcher has to undertake research using participatory methods such as Seasonal Calendars and Gender Daily Calendar. Seasonality calendars can cover themes such as water availability, disease occurrence, patterns of household food security (e.g maximum food availability and food stress periods), labour demand and so forth. Daily activity charts show people's routines and chores from morning to evening during both dry and rainy seasons while seasonality in terms of disease occurrences may cover both animals and humans (bilharzias, cholera, malaria and measles). These are key ingredients in, for instance, preparing a community's action plan. What is important is to be aware of the moments when it is opportune to do research without causing severe disruptions to daily chores in their lives.

Communication tools
Another challenge that a researcher has to take into account is the issue of "communication tools" that one uses. Whereas in western countries, they can easily make use of the internet, mobile phones and so forth. This can be totally different in developing African countries as most successful interviews have to be done face to face. Financial resources and equipment are a huge problem for African researchers. They sometimes cannot afford to buy the photographic, Personal Data Assistant (PDA), audio and video equipment to undertake their research. Thus, the approach of using technology as a means of doing research is definitely different in an African context as opposed to a western one. In the West, researchers have a wider access to the internet which can be used to dispatch questionnaires and the PDA, particularly in the Health sector, has been found to be useful with potential for quick turnaround times between data entry and processing and reporting. Overall, their use has enhanced the quality of the data.

Power dynamics
African and western cultures show differences in the manner in which they open up to strangers/researcher. In an African setting, there are issues which one cannot discuss with a stranger/researcher which implies that respondents will withhold information making it difficult to collect data. In an African setting, it is extremely difficult to have access to respondents, wife and children, without the explicit permission of the head of the household who is usually male (in a patriarchal society). Therefore, culture is another challenge that one would have to contend with when using western methodologies. As an example, the head of a household in most African traditional set ups is a male and it is a male who normally speaks to a stranger -in this case a researcher. This means that gender dynamics are also another challenge which implies that a researcher cannot talk to or interview a woman alone in the absence of the husband or a male relative. All these situations tend to influence the type of responses that a researcher gets. In other words, these challenges have a bearing on the outcomes of the research in an African context. In essence, it seems there is enjoyment of privilege by males that power and authority bequeaths upon them because of their gender in an African context. This would be interpreted from a western perspective as a reproduction of inequities when the purveyors of this custom view it as normal. Gender stereotyping is also commonplace. In a place like Gwanda in Zimbabwe, one cannot talk to the wife without the approval of the husband. (In some places, men and women do not use the same toilets). This implies that researchers ought to be culturally, politically and contextually sensitive as they do their research in any given setting. Samples at the individual level are not always easy to build in African context -for example a woman is less likely to take questions in the absence of their husbands if they are married due, again, to the power dynamics at play. This is different from the West, for example, in UK or the Netherlands, women are free to take questions and in most cases they do not need the approval of a man to speak their mind or to answer questions. The power dynamics between males and females, between genders -between males and males; and females and females -are at the core of the melee for equity in the distribution of power, opportunity, authority, self-esteem and resources between traditionalists/cultural conservatives and feminists.

Accessing respondents
Access can also be limited because one has to consult the traditional authorities who use their discretion to grant the researcher the permission to canvas information from his/her subjects. Entry has to be negotiated. In some rural areas of Zimbabwe, for example, a family may not easily entertain "strangers" to come into their midst without the knowledge and sometimes approval of the Chief/gatekeeper. This means that the Chief needs to be informed of the research and its purpose before it is undertaken. He can either refuse entry or approve it. He can also disapprove the subject to be discussed, for example if it threatens their cultural values. Traditional authorities can thus make or break a survey or study. It is no wonder that people can boycott the study if the chief does not approve of the study. Traditional protocol has to be observed. It is advisable to know the ethics and norms of the people one is going to research and know how to talk and address them without offence. Participation is not always easy to secure in some contexts. The challenge is also that how do you assemble people without paying them, if the intention is not to compensate. Concomitant to this is the challenge of assembling people at the same time where they have to abandon their usual chores for this data collection event.
The following example shows the span of authority of the Chief. The third national census of South Africa since the attainment of democracy in 1994, Census 2011, took place on 10 th -31 st October, 2011. In terms of the Statistics Act of South Africa, it is obligatory for citizens to take part in the census. It was reported on National TV that a Chief, in the KwazuluNatal province, had refused entry to census enumerators because his subjects had not been offered the first opportunity to work as enumerators. Given the importance of the exercise to the country and after elaborate negotiations and concessions with the chief, his subjects were eventually enumerated. This illustrates the power and authority that chiefs/gatekeepers command in an African context. Physical access to respondents can be impaired by the impassable roads, the forests and mountainous terrain that the researcher has to navigate. A researcher would need to establish a sample frame in order to draw a sample in such situations. There are a number of problems that a researcher would face in an African setting. For example, house numbers are found mainly in urban set ups and not in rural areas. It would be difficult to draw up a sample of houses in rural areas because the homes are not serially numbered or sequentially ordered and usually not located in a linear fashion (and often cyclical). It becomes a challenge to employ probability sampling methods such as systematic sampling in an African context. This is because the enumerator must cover all units in some sequence either even or odd until the last number in the sequence. The homeless also present their own challenges because of their homelessness and lack of fixed place of abode. The informal housing settlements in urban areas also pose their own problems as numbering is not sequential if at all it exists.

A case study based on my own experiences
I did a study which focused on the perceptions of poverty in Mhondoro communal area which is located in Mashonaland West province of Zimbabwe.
While the abiding theme of my work was poverty alleviation and targeting in poor African countries such as Zimbabwe, the hypothesis central to my study was stated as follows: Hypothesis: While income may be an aspect of poverty in Africa, access to assets is important in a self-definition of poverty that uses local norms as the measure of well-being.
Accordingly, the other questions that were subjected to interrogation were:


What factors affect perceptions of poverty and thus should be taken into account in a consensual approach to poverty?  What is the minimally adequate asset level (MAAL) that a person must have for him or her not to be considered poor?  Is there a measurable threshold of assets that constitutes poverty?  How many and what kind of assets do people need not to be poor?  Does location matter?

Aims of my study
The study had one overarching objective: To establish an asset-threshold using the consensual approach through the eyes and experiences of the people of Mashonaland West in a self-definition of poverty which uses local norms as the measure of well-being.

The research setting
Mhondoro communal area is situated 40 kilometres from Harare, the Capital City of Zimbabwe. Villagers in the area grow plants and keep animals for day-to-day living. This communal area was chosen, firstly, because the area reflects a social landscape that is rural -this was one of the fundamental conditions underlying the quest for an assetthreshold steeped in a rural setting. Secondly, the researcher's familiarity with the location as well as knowledge of the local language, Shona, which is critical in understanding the perceptions of poverty and how these relate to assets. Shona is the dominant language spoken in this province. Familiarity with the location brings in some element of purposiveness in terms of site selection. About 66 per cent of the population had no electricity while 80 per cent of the population had access to safe water, either from piped water or from boreholes/protected wells. The remaining 20 per cent obtained water from unsafe sources such as unprotected wells, rivers and streams.
To gain entry I also had to obtain the permission of the kraal head despite the fact my father was born in that village. It was not a given that I would be allowed to collect data for my study without his permission. Since I was staying in the capital city and had last gone there many years back, my brother had to consult the kraal herd before I could do my study there. The kraal head knew my brother because my brother has built a house in the village and he commutes there every two weeks. He also has a small plot adjacent to his house on which he grows maize and vegetables. After negotiations on my behalf, permission was granted. In a way I felt like an outsider/stranger. However, after the permission was granted, work commenced. It was easy for the kraal head to assemble his subjects. He would hit with a heavy metal rod for about six to seven times onto a thick metal sheet shaped like a satellite dish which hung from a tree, which sounded to me like a Church bell, to call his subjects. In a few minutes, the villagers would start streaming from their roundavels and houses to the kraal head's compound. He announced the object of my visit and granted permission to his subjects to assist me with the information I needed for my study. With the kraal head's blessing, I had more households from whom I could canvass for information than I had anticipated. In terms of research design, I used the ballot (or poll or voting) in which people were given ballot papers as is done during parliamentary and presidential elections. Instead of voting for their member of parliament or president, they voted for the assets they deemed to be necessary to achieve an acceptable standard of living. All participants in the poll were given a unique number which was used to link their ballot papers to their demographic data. Each ballot paper had the name and picture of each asset. Three categories of ballot boxes were used for this purpose: Most Important, Important and Less Important. Accordingly, in the process of voting, participants were asked to place their ballot papers in the appropriate box. After the votes were cast, the ballots were counted before the assembly of villagers for validation.
Six focus groups in the rural areas and 3 in the urban area as well as interviews complemented the vote in this multi-method approach which collected both quantitative and qualitative data. Interviews were carried out covering demographic information as well as definitions of poor and rich persons. The first choice participants for Focus Group Discussions (FGDs) were heads of household. In their absence any member of the community participated because assets are 'neutral'. If the concern was income, the researcher would have involved primarily only income earners in the FGDs. During the main survey, six FGDs were held in the rural area: one adult male-only group; one adult female-only group; one adult mixed group; one youth-male group; and one youth-female group and one mixed youth group. In the urban area, only three FGDs were conducted as follows: one adult male-only group; one adult female-only group; one mixed youth group. Assets concerned those who had them as well as those who did not. On average, between 8 and 15 participants took part in the FGDs.
There were separate FGDs for males and females. The purpose of separating the males from the females was to allow the different genders to speak their minds freely. It was also of interest to delve into the different perspectives that men and women had on a subject and separation of genders allowed this. Men and women may have different perspectives on say a house. Men may need a house, but women may look beyond a mere house to its quality as well. Males and females are not homogenous social categories in terms of needs and aspirations. Mixed groups of both males and females were also used to affirm views, to engage men and women in debate (if there was such a debate) and to establish dominant views in constructive engagement. The mixed group FGD dealt with the issues of how many of the items they voted for were required to lead a minimally acceptable way of life as well as to determine the cost of each item.
Accordingly, FGDs were useful in the clarification and elaboration of social norms, experiences and practices; and providing a broad overview of perceptions of what constituted assets that were necessities of life for the people in the area. In the rural area, the list consisted of 32 items, while in the urban area it contained 37 items. These items were chosen by participants in FGDs after starting off with a short-list of 14 items. The preliminary list used in the urban and rural areas had the following 14 items: cattle, hut, land, plough, wheel-barrow, scotch cart, hoe, goats, spade, axe, bed, chickens, radio and donkey. In the urban area, the list excluded a cultivator and harrow mentioned in the rural area and also separated a mobile phone from a telephone as distinct assets. The people selected these items as a broad yardstick by which the poor could assess their lifestyle.
The consensual approach is where public opinion is used via a ballot to determine socially perceived necessities of life. (See Mack and Lansley (1985) who are the proponents of the methodology used in this study). Middleton (2000) notes that consensus can also be reached through discussion, negotiation and eventual agreement. Thus, the lack of certain goods and services which are believed by the majority of the population to be essential falls under consensual measures (ibid).
Because of the success of this first leg of my study, following the rapport that I had established with the kraal head and his subjects, I followed it up with another study in 2009. I have turned my study into longitudinal study. This case study was sourced from Mtapuri, (2008Mtapuri, ( , 2010.

Workshops and seminars
Workshops and seminars are a form of data collection. Facilitators will solicit ideas, opinions and views from different and disparate stakeholders on a matter. Brainstorming can be used to gather information. The flow of information should be encouraged back and forth, unfettered. A trained facilitator can make that possible. Rich and useful information can be obtained for further processing from these open discussions. The advantages are many. For example, stakeholders will have an open opportunity to air their views. Workshops and seminars usually come at a great cost. These costs include hiring a venue and payment to participants and facilitators. Because of paucity of resources in an African context, workshops and seminars can be held under a tree, in a marquee, tent, a classroom, and so forth.

Open mic (microphone) sessions
Open mic sessions are ideal for the youth (especially in urban areas). These are sessions in which everyone has a chance to air his/her views in front of everyone in an open and transparent manner. It allows people to vent their anger, emotions, joy and happiness at the issues that matter most to them.

Storytelling
There variants to storytelling. One of them involves sensitizing respondents on the theme for discussion so that the researcher and the respondent are on the same wave length. The researcher then asks the respondent what they know about the issue. He/she further probes as the conversation continues. Alternatively, an imaginary story is presented and a respondent is asked to relate it to their own experience. Second variant is for people to form a group. And then let each person tell their own story which is recorded on tape with their permission but recorded anonymously. By giving them the tape recorder, the researcher will be giving them control over what they say. The tape recorder is passed on like a baton stick to the next person who also tells his/her story. The third variant is to engage with the respondents one-on-one and the researcher then asks a specific question to kick start the conversation. From then on the respondent will tell his/her story. The fourth variant is done without giving any direct prompt. The researcher captures stories spoken when people are busy with their chores for instance baking or stories spoken as people gossip. It assumes an open discussion. No topics is pre-determined. It is conversation between people who know each other. It may be about their individual aspirations, community aspirations, politics, sexuality, infidelity and so forth.

Event search
This approach to data collection which I call event search is a way to understand how people deal with life experiences. It seeks to gather information on their ways of life at key events in people's lives. These events include weddings, marriages, ritual ceremonies, funerals and so forth. It is these events that speak to the real life experiences as they reminisce about themselves, friends, workmates, their rulers and leaders and their communities in general. These are the stories that shape and point to the way people live their lives. One will not be interested in the wedding or funeral per se but about what people say and talk about in relation to socio-political issues such as housing, electricity, sanitation, life in general and so forth. It is at funerals that people talk about HIV/AIDS and sexuality and even laugh at some of the issues.

The etiquette challenges
What would a researcher do if he/she is offered water to drink when he/she can see that the source of the water is dirty or the water does not meet his/her hygiene standards? In rural areas of Zimbabwe, people drink traditional beer from the same gourd -a gourd is a hardskinned fruit which is dried and hollowed out to make a bowl or cup -which rotates from one drinker to another without being washed. Skipping is grimaced and frowned upon.
What would a researcher do if offered a gourd of beer/water that is in rotation? Some may consider it unhygienic while beer has been traditionally drunk that way from generation to generation. In Rushinga in Zimbabwe, the people smoke marijuana openly. Even children are allowed to smoke marijuana in this region. However, in the rest of the country it is illegal to smoke marijuana. What would a researcher do if he/she observes an illegal activity during his/her research and children are partaking in it? This poses an ethical challenge. In the same region, beer is brewed in a rotational fashion from one household to the next on a weekly basis as an income generation venture and a practice. Anecdotal evidence shows that women in the region even drink more than men because it is the responsibility of the women to brew the beer. This has been the case since time immemorial. In such circumstances the researcher must not judge the subject and colour or taint his/her research with his/her own biases.  Table 2 shows the opportunities and challenges that a researcher can anticipate in an African context. Opportunities are galore in the realm of poverty, unemployment, rituals and customs, gender as well as a myriad of other social and economic issues, including the environment.

Opportunities
Challenges Learn African rituals and customs e.g initiation ceremonies The customs are considered sacred and opportunities are lost to study them. Using hidden cameras becomes unethical. Learn about gender and gender imbalances Values and beliefs embedded in cultures and passed on from generation to generation To use photo narrative -in which you use photos to prompt discussion. People say what they see in the photo and asked to draw parallels with their own experiences.
Technology is not readily available and not affordable to many people in the rural areas To use overt observation Access is regulated by the chief, kraal head or male members of the family To appreciate and celebrate diversity The existence of many languages means that meaning may be lost during translation from vernacular to English and vice versa. Diverse groups keep their customs, beliefs and traditions secret

Conclusion
In this Chapter I presented challenges of doing research in an African context. There are differences in how issues can be conceived in an African as opposed to a Western perspective. The differences could be cultural/traditional, literacy levels or of a moral nature. This does not imply that doing research in Africa is fruitless or impossible but rather the Chapter warns the researcher of what to heed and pay attention to. It was also about presenting a worldview that thrives in an African context so that the researcher is better prepared to navigate his/her way in that space. People do successful research in an African context by findings ways around those apparent and fundamentally technical ambiguities presented by Western and African perspectives. To be a successful researcher in Africa and perhaps elsewhere, it requires that a researcher is culturally, politically and socially sensitive in the environment in which one is doing his/her investigations. This is helpful in appreciating what permissions need to be obtained to gain entry into communities, what methods are ideal for data collection in the circumstances, who should be the subject of the investigation, how would one draw a sample and what sampling method should one choose and so forth.
In this chapter we exposed the challenges found in an African context such as:


The chief and the headman becoming important role players in surveys because without whose consent, access to subjects may be denied. The chief may approve or not approve the topic for the research;  In some patriarchal communities in Zimbabwe for example, a woman may not respond to questions in the presence of the husband. It is the husband first as his prerogative, which he can delegate to her.  Certain rituals like initiation are done in secrecy to preserve their "sacredness", which poses a challenge when doing research;  In places where respondents are illiterate, self administered questionnaires become redundant.  A team of multi-lingual research researchers and assistants may be required due to the availability of a wide array of languages thriving in African communities. Translation from vernacular to English and vice versa may not capture the essence of the discussion and hence meaning may be lost in the process.  When sampling, literature talks about numbering the houses, street by street in some sequence, for example, for systematic random sampling. But in some rural areas of Africa, such streets do not exist as roundavels/huts in a village are clustered in a manner that may defy sequential numbering.
The challenges highlighted above do present opportunities. It is for the avid and passionate researcher to navigate in this less understood but interesting and exciting space for research -the African terrain. The African terrain, while providing a seemingly contradictory and ambiguous tapestry when juxtaposed with a western one, it presents a diverse and rich terrain of unlimited possibilities for research. As my experience has shown, my snowball was growing as more and more people wanted to be involved in my study through word of mouth. This happened because I observed traditional protocol to the letter as well as all relevant research ethics governing my work. The sky is no longer the limit. I conclude with these perceptive words by Khotsa and Sithole: Africa has the potential to become a world leader in almost all spheres of life, but potential remains potential as long as it is not explored and exploited (Khosa, 2009:11).
Knowledge production, scientific knowledge production in particular, should continually be informed by discourse battles and debates, openness to innovation and discourses from elsewhere (Sithole, 2009:98).
I choose to take the opportunities and still imagine a rich and flourishing terrain for knowledge production in Africa.

Acknowledgement
I would like to thank Jana Sertic who managed the publication of this edition, my colleagues Pudurai Justin Mazengwa, Judy Mohale, Nompumelelo Thabethe, Mamoloko Rangongo and Osden Jokonya for their counsel and thoughtful suggestions during the preparation of this piece. I am grateful to Prudence Lebese and Maritjie Du Toit for their support during the data collection and analysis process and my colleagues at the Turfloop Graduate School of Leadership of the University of Limpopo for their support and guidance throughout this project.