The impacts of COVID-19 have forced researchers to adapt quickly to working under unprecedented restrictions. Here we discuss a new toolkit for Remote Data Collection, gathering researchers’ solutions on everything from replacing face-to-face interaction to greater use of secondary data. The toolkit promises to be a useful resource for the research community, and a guide to improving the efficiency of data collection.
Mindel van de Laar: Over the last few months, with a team of four researchers, I created a toolkit on ‘Remote Data Collection’, which includes information on a range of topics that became an important part of research during COVID-19 — when data collection practices were often no longer feasible in person. It was created as an immediate response to requests by our PhD candidates on how to best deal with online data collection. The content of the toolkit is based on our own experiences in conducting remote data collection. While it is certainly not complete, the toolkit offers guidelines and practical advice for collecting data online.
Talitha Dubow: Indeed, COVID-19 did affect our research practice and we had to respond with new processes as a research team. Just as the first lockdowns were being implemented, I was due to go to Iraq and Senegal to conduct fieldwork for Work Package 2 of the ADMIGOV project. When it became clear that international travel would not be possible for the meantime, we adapted our planned fieldwork for implementation by local research consultants in Iraq and Senegal. These adaptations were specific to each country context.
In Senegal, where in-person data collection was still possible, a research team led by Mamadou Dimé at the University Gaston-Berger of Saint Louis conducted semi-structured interviews in-person. For these interviews, we slightly modified the interview guide in order to better ensure a consistent approach across the team of 20 fieldworkers. These interviewers also kindly provided me with their own fieldwork observations and, in some cases, photographs, to aid my understanding of the local context and interviews.
In the case of Iraq, all interviews were conducted by a local consultant, Botan Sharbazheri. However, these interviews had to be conducted by telephone because of the greater health risks. For these interviews, we had to shorten the interview guide in order not to fatigue interviewees with long telephone conversations, and we had to start the interview process with some additional questions designed to get a better sense of the interviewee’s environment and their comfort within this environment. These revisions were made in order to make up for the researcher’s lack of ability to see the research participant’s interview setting and body language. I have really enjoyed collaborating with the local researchers in both countries, and the research has greatly benefitted from their ability to access hard-to-reach groups, their sensitivity and skill in navigating the risks and challenges of the data collection process, and their in-depth understanding of the local context and dynamics.
Mindel van de Laar: Several of the physical interaction moments that normally occur when collecting data, now happen virtually with study participants and researchers physically distanced. In Unit 5 we focused specifically on those interactions, and the need to connect with your participants explicitly. This unit is based on our own experiences. It is based on these understandings that we identify some of the main difficulties faced by our researchers in recruiting and interviewing people online.
Soha Youssef: A good example is when I was working on two projects when we suddenly had to transition to online fieldwork. I was actually set to travel to Iraq and Nigeria, and everything was ready but then we had to find a way to conduct all the fieldwork online particularly for the project “Connecting Diaspora for Development.” Difficulties already started with how to recruit key stakeholders. When we conduct in person fieldwork it is easier to have snowball sampling. When you are present in the community and have access to the target group, reaching out is less challenging.
The biggest difficulty in conducting virtual interviews, is internet connectivity. The key stakeholder recruited needs to have access to stable internet connection and also have access to the platform where we are conducting the interview. This can be a big challenge to many key stakeholders, resulting in interruptions during the call, delays, rescheduling, or no show of the scheduled key stakeholder’s interviews. I personally experienced a higher rate of no-show. When you are in the country, you can often walk from one interview location and interviewee to another, and in person presence reminds interviewees about their commitment. When you are interviewing online, at best I can sit down at my computer waiting for the person to show up. In case people do not join the call, there is not much I could do.
I also experienced that establishing trust with the key stakeholders was a bit harder when conducting interviews online. In person you both see each other and relate directly. Virtually, when you switch on the camera you may compromise the stability of the internet connection. Often the result was an interview without visual interaction. Yet, overall, I was able to find solutions for the difficulties I came across and in unit 5 I give all the recommendations and the suggestions based on what have worked for me.
Mindel van de Laar: Whilst potentially less obvious, in the use of secondary data we experienced obstacles in reviewing data fit and accessing secondary data online. Researchers cannot visit institutions in person, and contractual agreements need to be dealt with remotely. When scanning several main data providers sites, we observed differences in the process of accessing secondary data.
Mathias Weidinger: Secondary data access varies across data repositories and who provides them. International organisations like the UN, World Bank, or OECD tend to allow unrestricted access to their depositories, but their datasets are sometimes messy or superficial. The UK Data Service, for example, has a slightly more restrictive approach in that users need to declare the purpose of their query and register yearly with the portal. In exchange, the data available on that portal is much more diverse.
There are also national statistical offices such as the Dutch CBS. They usually offer a limited amount of high-level data for open access. Whoever wants to dive into their massive pool of microdata needs to go through a rather long process that usually involves personal meetings with the responsible staff members. Under the current circumstances, however, all necessary steps can be taken remotely. The added scrutiny of CBS and similar institutions is warranted as their microdata may include pieces of information that, taken together, could identify individual participants.
Quantitative data is much more abundantly available than qualitative data, which is perhaps due to privacy concerns – long interview answers can give away much more personal information than your average survey datapoint. Nevertheless, I also found a portal specifically dedicated to hosting openly available qualitative data, which is hosted by the University of Syracuse, NY.
With a little digging, there is a wealth of data quite easily available on the internet. So before spending time and effort on collecting your own new dataset, it might be worthwhile to try and find existing data that suits your project.
Mindel van de Laar: Thinking of e-resilience in data collection, COVID-19 has pushed research societies into using remote data collection more frequently. In doing so, we increased our capacities and understanding of how to improve data collection, and how to collect better data using remote tools. Our ‘forced’ journey into remote data collection may benefit the future of data collection exercises, even after we have dealt with the current COVID-19 pandemic.
Talitha Dubow: Firstly, I think that, for many of us, the pandemic context has prompted us to engage with a wider range of research methods and approaches beyond those that we are most familiar with. Personally, I have been very excited to learn more about what other researchers have been doing, both before and during the pandemic. I think that, as a research community, we have had to pay greater attention to the feasibility, ethics, advantages, and drawbacks of different remote data collection approaches, and I hope that there will be widespread efforts to take stock of what we have learned from this past year (or more) of adaptations and innovations in data collection strategies. In this regard, it also seems that, as researchers, we have become more accustomed to attending online webinars, and I hope that the online option for participation continues in the post-pandemic world, because I think that it has been a fantastic way to learn from other researchers around the world, without the time- and financial costs of international travel. Finally, and as described above, I am also very glad to have had the opportunity to collaborate with local researchers in Iraq and Senegal, and I would like to build on this experience in order to develop productive and equitable research collaborations in my future work.
The opinions expressed here are the authors’ own; they do not necessarily reflect the views of UNU.
Pexels / T. Miroshnichenko