Fieldwork experiences can often be a daunting way of conducting research but they can also be fulfilling. I have had first-hand experience conducting fieldwork for my master’s degree, and while it was a generally enjoyable experience, I did make a number of mistakes during the process. However, I learned some valuable lessons as a result of this too. During my fieldwork in Rwanda, I increasingly realised that it was important to incorporate primary research data into my study, but because of a lack of data on my topic, I made use of other sources of qualitative data to validate my findings. This strategy, according to Denzin (1970), is known as methodological triangulation and it allows researchers to make use of various data gathering methods to ensure internal validity. Based on the use of methodological triangulation, I specifically designed interviews targeted at both elite groups and slum dwellers in Rwanda to investigate the thinking behind the urban policies designed by political elites, and how it impacts marginalised slum dwellers.

The first group of interviewees that I targeted comprised of government officials and the second comprised of a group people living in slums. The elite interviews were generally semi-structured in nature and were based on open- and closed-ended questions. Scholars such as Harvey (2011) have noted that this is the best approach for elite interviews because it allows flexibility and hence, maximises response rates. Notably, scholars such as Aberbach and Rockman (2002), Hoffmann-Lange (1987) as well as Zuckerman (1972) have also shown that elites prefer to engage with open-ended questions so that they can articulate their views coherently. During my interviews with elite groups, I did not always draft formal questions, but I made sure that I was familiar with the topic so that I could comfortably develop a natural rapport with the respondents. Interview times ranged from thirty minutes to up to two hours, and respondents commonly offered me some data sets to consult, which followed with some discussion. I was often required to submit a formal application to access these files, and although I formally submitted a request letter to the Ministry of Urban Development, my efforts proved futile, and it became increasingly frustrating for me to access the data sets I needed.

In some of the interviews, I found that political elites provided evasive answers because of the politically sensitive nature of some of the questions posed. The general etiquette according to Peabody et al. (1990) suggests that political elites should ideally not be interviewed using recording devices as it can cause elusiveness and anxiety – although I did not use a recorder, I continued to receive elusive responses which sometimes left me frustrated and disillusioned with my project. I felt particularly irritated because the absence of a recording device meant I was unable to get hold of a verbatim record of my interviews. Because I had to write down observational notes while engaging with the respondent, it was difficult to record all the information and I lost out on some important points. I tried to strike a balance between note taking and the interview process, but I found this to be a difficult endeavour. I was able to access more political elites than initially anticipated, however it often felt futile because I couldn’t source as much information as I had wanted from this sample group. I tried to counteract these limitations by shifting my focus to the second sample in my study, the slum dwellers, although this was also fraught with some complications. Comparatively and overall, the second sample group proved to be more cooperative and I quickly learned that I had wasted a significant amount of time focusing on political elites, when a lot of the responses I desired could have easily been sourced from policy documents and government reports.

As mentioned, I discovered that slum dwellers, after gaining their trust, provided a great deal of nuanced insight into my understanding of urban regeneration in Rwanda, which was very beneficial for my project. Harvey (2011) has highlighted how field researchers must endeavour to earn the trust of their respondents to gain access to high quality data and looking at the results I garnered, I believe I was able to do this successfully. The data acquisition from the sample group was however, not without complications. The first complication pertained to my status as a foreigner, which I realised made several people wary of my presence. After visiting the research site continuously over a period of time, they became more familiar with me and thus opened up to the idea of participating in my study. I also ensured that I hired a local research assistant, and I realised that my association with a local gave me a greater deal of legitimacy in the eyes of my potential research respondents.

While oftentimes the data collection process was extremely stressful, and sometimes precarious, I learned to be resilient in, and how to maintain focus on meeting my set objectives. Concurrently, I also learned when to change approaches in the field – especially when a particular research method had proven to be unsuccessful. In hindsight, I should have changed my approach much earlier to save a lot of the time I wasted. Looking back, I would have placed less emphasis on the elite sample group as primary data was not necessary for addressing my research questions concerning government policy. I could have saved time and effort in sourcing this information from secondary sources such as government reports and books. I also would have employed a local researcher much earlier in the process as it paved the way for gaining the trust of respondents. At the same time, I realised that I should have provided a lot more training for the research assistant who also served as a translator, due to the events that ensued in the field.

According to scholars such as Temple and Edwards (2002, p.2) “the interpreter is a conduit linking the interviewer with the interviewee and ideally is a neutral party who should not add or subtract from what the primary parties communicate to each other” but in my research, I quickly realised that this was not the case. Generally, the research assistant was highly opinionated and in some instances tried to impose his political views on my respondents. Looking back, I think I could have done a better job in training him and getting him to understand why the responses of interviewees should not be pre-empted. There were several instances where I also noticed that the translation process was not as effective as it should be during the fieldwork process. Due to my increasing familiarity with local dialects, I was able to discern when the translator was not providing the full picture with respect to the responses of the respondents. In my opinion, this was indicative of the lack of training which the translator received and I learned to not just assume that job roles were obvious, especially in this context. In instances where omissions were obvious, I questioned the translator to gain further details. During the early stages of the fieldwork, he was also far too independent and in some instances, did not stick to the script, in terms of the interview questions I had drafted. In hindsight, I should have done a trial run or pilot study, so that he was better acquainted with the standard of research I was expecting.

Overall, the process was a challenging one that introduced me to the iterative nature of fieldwork. It became increasingly obvious to me that in the field, nothing ever goes as neatly as planned on paper. I realised how imperative both resilience and flexibility was in the field. In hindsight, I would have prepared back-up alternatives for each aspect of my study, since in some cases I was completely thrown off guard and had to take a few days off to re-strategise. I lost a few valuable days by doing this and if I had managed my expectations with more caution, I believe I would have been better prepared for the unexpected occurrences in the field. On reflection, I would also have taken my positionality far more seriously, as I never imagined that the way I was perceived by others could affect my study. During my interviews with elite groups, they were often bemused and yet intrigued at the same time about my role as a foreign female researcher. Scholars such as Kobayashi (1994) have highlighted how gender identities play out during fieldwork, meaning women are often discriminated against on the basis of their sex. I did not experience discrimination, but I feel my being a woman helped me gain access to certain respondents that I never anticipated to access, because I was somewhat perceived as a damsel in distress who needed help. While the pity I received worked in my favour, it had no bearing within the interviews themselves as I was not able to acquire the data that I hoped to. Having a better understanding of the cultural nuances would have helped me to manage my expectations better. Indeed, researchers such as Denzin and Lincoln (2011) have highlighted how a researcher’s ability to gain access is shaped by personal characteristics, including gender. Looking back, I believe my fieldwork project was fairly successful, mostly because of its ability to gain nuanced insight from the second sample involving slum dwellers. The major pitfalls of the project mostly pertained to the logistics of the project – specifically the lack of training for the research assistant, and the general lack of a research focus.

References
Aberbach, J. & Rockman, B., 2002. Conducting and Coding Elite Interviews. PS: Political Science and Politics, 35(4), pp.673–676. Available at: http://www.jstor.org/stable/1554807?seq=1#page_scan_tab_contents [Accessed November 20, 2015].

Denzin, N., 1970. The Research Act in Sociology, Chicago: Aldine.
Denzin, N.K. and Lincoln, Y.S., 2011. The Sage handbook of qualitative research. Sage.

Harvey, W.S., 2011. Strategies for conducting elite interviews. Qualitative Research, 11(4), pp.431– 441. Available at: http://qrj.sagepub.com/content/11/4/431.abstract [Accessed May 6, 2015].

Herod, A., 1999. Reflections on interviewing foreign elites: praxis, positionality, validity, and the cult of the insider. Geoforum, 30(4), pp.313-327.

Highley, J., Deacon, D. & Smart, D., 1979. Elites in Australia, London: Routledge.

Hoffmann-Lange, U., 1987. Surveying national elites in the Federal Republic of Germany. In G. Moyser & M. Wagstaffe, eds. Research Methods for Elite Studies. London: Allen & Unwin, pp. 27– 47.

Johl, S.K. and Renganathan, S., 2010. Strategies for gaining access in doing fieldwork: Reflection of two researchers. The Electronic Journal of Business Research Methods, 8(1), pp.42-50.

Kobayashi, A., 1994. Coloring the field: Gender,“race,” and the politics of fieldwork. The Professional Geographer, 46(1), pp.73-80.

Mullings, B., 1999. Insider or outsider, both or neither: some dilemmas of interviewing in a cross-cultural setting. Geoforum, 30(4), pp.337-350.

Nolinske, T., 1995. Multiple mentoring relationships facilitate learning during fieldwork. American Journal of Occupational Therapy, 49(1), pp.39-43.

Rose, G., 1997. Situating knowledges: positionality, reflexivities and other tactics. Progress in human geography, 21(3), pp.305-320.

Overing, J., 1987. Translation as a creative process: The power of the name. Comparative anthropology, 70, p.71.

Peabody, R., Hammond, S., Torcom, J., Brown, L., Thompson, C. & Kolodny, R., 1990. Interviewing Political Elites. PS: Political Science and Politics, 23(3), pp.451–455.

Temple, B. and Edwards, R., 2002. Interpreters/translators and cross-language research: Reflexivity and border crossings. International Journal of Qualitative Methods, 1(2), pp.1-12.

Zuckerman, H., 1972. Interviewing an Ultra-Elite. Public Opinion Quarterly, 36(2), p.159. Available at: http://poq.oxfordjournals.org/content/36/2/159.abstract [Accessed November 20, 2015].