[DATA COLLECTION REFLECTION- SARITA PANDAY] A reflection on managing challenges related to data collection in Nepal

Nepal has achieved significant improvements in its maternal health in the last decade and has extensively mobilised community health workers known as Female Community Health Volunteers (FCHVs) across the nation. FCHVs are the first source of contact for women requiring Maternal Healthcare Services (MHS) in the village. They have been widely utilised by both public and private health centres in almost all the community health programs. While there is increasing responsibility on FCHVs to solve the problem of access to MHS such as antenatal care, referral services, adequate attention has not been given to study the experiences of FCHVs themselves. My research explores the experiences and perspectives of FCHVs in MHS delivery and those of mothers, expectant mothers and local health workers using interviews and focus group discussions. Interviewing different groups of people related to the FCHVs’ work was invaluable in interpreting how FCHVs perceived their work. I collected data from two different parts of Nepal from April to September 2014. I came across several methodological challenges during the study which I am going to address in the following paragraphs. I will particularly highlight the issues related to travelling to the data collection sites, use of a local guide, reflecting on my own position and managing interruptions during the interviews.

Travelling to the data collection site was full of challenges: due to the difficult terrain, poorly built roads and low quality transportation, I had to walk for five hours to reach the first data collection site. The roads in the remote villages were dusty during hot days and muddy on wet days. Transportation was also considered to be a major problem by my research participants in hills, as difficulties in travelling led to limited access to MHS in the region. Travelling to the second place, the flat lands of Nepal required eight-hours by jeep through one of the most dangerous routes from the capital city Kathmandu. In most cases, I managed to reach the interviewees’ houses but in a few instances, interviews and group discussions were rearranged by mobile phone at places convenient for both the researcher and the interviewees.

Locating possible participants and planning for interview dates were preliminary steps to the data collection. I had meetings with health workers in the public healthcare centres where FCHVs gathered for their monthly report submission or training. Attending these meetings enabled me to meet and set interview dates with both health workers and FCHVs. I was also 1

[DATA COLLECTION REFLECTION- SARITA PANDAY] able to see how health workers behaved with FCHVs and how FCHVs viewed the support gained from health workers. Understandings of different perspectives were invaluable to interpret how FCHVs, health workers and women were responding during the interviews. For example, some FCHVs expressed their concerns about health workers not being supportive to them. I myself could see observe and record the power imbalances between health workers and FCHVs as some health workers gave direct orders to FCHVs without giving them due respect.

In the data collection sites, support from local guides remained vital for undertaking interviews and group discussion. I was new to most of these areas and some parts of the village were far from the main centre. A local guide accompanied me to the interview sites. S/he introduced me with community members and the research participants thus helping me to build rapport with them. Without the presence of a local guide, it would have been difficult to get information on MHS as illustrated by one attempt to collect data on my own. I identified a suitable woman for my interview by asking people in one village. However, I realised her husband was reluctant to allow her to provide information as I was unknown to them. When I responded to all his queries, I was able to interview the woman. In other cases, having a local guide from the same community helped me to communicate more easily.

I had to be reflective every moment while accessing MHS information from my participants as the people’s preconception about me could affect the quality of data. For example, I had to think about how my background, my thoughts and the interview process could affect the quality of MHS information that I needed. My ethnicity was different from the members of the community and these affected perceptions. I realised how FCHVs’ perceptions of me changed their response in some interviews and group discussions. While I was younger compared to most FCHVs; because of educational differences some FCHVs felt that I was superior to them. This was clear in one group discussion and in some interviews when FCHVs reported only positive aspects of the job at the beginning. However, when I emphasised that I was there to listen to them and their concerns on both positive and negatives aspects of their volunteer work, they started explaining their issues more clearly.

Regarding the interview process, my own naïvetés in the qualitative interviews caused difficulties in eliciting enough information in the beginning. I found it hard to convince women to talk more despite several attempts to ask further questions through probing. 2

[DATA COLLECTION REFLECTION- SARITA PANDAY] Women tended to talk less with frequent short responses. Shyness could be one possible reason for this. Another reason could be because they women had never been exposed to research interviews before.

Managing interruptions during the interview was challenging, but was important for the data quality. I had to arrange interviews in a place and time offered by the participants as their participation was voluntary. Most of these interviews encountered some interruptions and disturbances from the participant’s husbands or mothers-in-law or neighbours as the interviews were generally held on verandas outside houses. Sometimes a curious passer-by interrupted the interview. Often I had to stop the recorder and inform the passer-by about the interview process and its significance while assuring them that I would get back to them as soon as I finished the interview. In the Nepali context, it is socially acceptable to ask questions or show interest in someone even if one hardly knows the other person. This is especially common if there is any new person in the village. I tried to minimise these disturbances by interviewing FCHVs or women in a separate room or a bit further away from their homes in the field or kitchen garden or nearby shops where they could talk comfortably without any kinds of pressure. In another instance, I interviewed a health worker in the restaurant next to highway. I didn’t realise how disturbing the background noise could be until later: some of the excerpts were incomprehensible. So, I became careful in the next interviews.

In this article, I explained how I interviewed my participants while dealing with issues such as travelling to the data collection sites, using local guides to access to participants, reflecting on my own position and managing interruptions during data collection. It was necessary to be flexible and cautious while accessing the data as every moment in the field revealed some important elements that were not obvious at first sight. I realised that even the best prepared plan might not work in the actual data collection site. Therefore, the researcher should be flexible enough to modify the plan as per the contextual needs while dealing with any issues that may arise unexpectedly.

3