by Scott Timcke and Zara Schroeder
During the course of its studies of the drivers of information disorders in African situations, Research ICT Africa has used a variety of methods and methodological approaches. Like Africa itself, our team views this pluralism as a strength to better comprehend the drivers of mis-, dis, and malinformation practices.
In our most recent research, we have begun to test and refine the utility of using WhatsApp as a responsive research method for micro studies. A micro study is a small-scale research study that is focused on a specific, narrow aspect or question within a broader topic or field of study. It typically involves a limited scope, sample size, and duration, and is designed to explore or investigate a particular issue or phenomenon in depth.
Since its inception in 2009, WhatsApp has become one of the most prevalent messaging apps. Taking advantage of this widespread presence and the app’s affordances, during the global coronavirus pandemic, researchers sought to creatively experiment with using the app to conduct qualitative research. For example, Scott used the app to study elite recruitment in Africa, their motivations, as well as these elites’ self-conception of science, technology, and innovation policy.
For the past two months, Zara has been conducting qualitative interviews via WhatsApp’s messaging feature. The main investigation of this micro study has been perceptions of online gender-based violence. This is part of our recent attention to how AI has amplified the scale and intensity of online gender-based violence, non-consensual intimate image distribution, and deep fake image manipulation to sexually degrade and exploit women. We are conducting research on these topics, examining the role of social media engagement algorithms to better equip African policymakers to participate in general and specific global AI governance discussions on these matters.
Given the sensitive nature of these topics, Zara has been very attentive to issues of data ethics, data handling, and the ethics of inquiring about direct experiences. For these reasons she has designed and confined the study to perceptions, conjectures, forecasting by users of platforms, and general attitudes, as opposed to seeking information about specific personal circumstances. This method also promotes a participatory governance approach to micro research as it enables participants to share their perceptions and contribute to ways in which policy and regulations can be developed to protect them online.
All method selection involves trade-offs. One benefit of using WhatsApp is that it has allowed our research team to asynchronously conduct semi-structured interviews with people from across South Africa. This geographic versatility is a finding that other researchers have repeatedly highlighted. As WhatsApp allows for the circulation of voice, text, emojis, videos and images, it affords participants a wide range of communicative expressions, allowing nuance and intertextual references.
As Ziyanda Mwanda explains, this “ease of use, convenience, and nativeness offers a range of modalities – although the most predominantly used are text and emojis. Voice notes are increasingly gaining popularity as they offer a quick response with less ambiguity. Text is words typed using the keypad of the participant’s device, which is mostly in the language of the user.” These varied forms of data invite other interpretative techniques that can enrich the analysis of meaning and implications. In this sense, WhatsApp data collection is generative for other layered techniques of inquiry, which and in turn lead to better policy formation.
While employing WhatsApp as a data collection tool offers several benefits, there are several limitations that warrant consideration. Given the sensitive nature of the research context, having participants share personal experiences via WhatsApp raises privacy concerns regarding access to their mobile devices. Although the app is encrypted, devices can be lost, thereby compromising conversations and the measures in place to safeguard participant’s personal data. For these and other reasons, we strongly believe that researchers should limit their studies on this medium and this app to perceptions. The study of direct experiences is perhaps best undertaken via other methods and modes.
Another limitation is the requirement for both participants and researchers to have access to a mobile phone and periodic access to an internet connection to participate in WhatsApp-based data collection. This prerequisite means some populations may not be covered. As such, this kind of app mediated qualitative method cannot claim representativeness or generalizability; however these are not the primary purposes of qualitative research or micro-studies.
Additionally, while the asynchronous nature of WhatsApp communication may be useful, it may hinder the flow of discourse and the ability to probe responses in real-time. Given their different geographic places, researchers must be attentive to context collapse, for example. Jennifer Loh and Michael Walsh explain this phenomenon arising from (online) content that “can speak to multiple invisible audiences who exist in different time zones, different places, and who possess different norms. Consequently, context collapse can flatten multiple audiences into a single context, making the management of the self and our online identities across varied settings increasingly complicated.” Researchers must consider these limitations when designing studies or analysing data.
In our view, there is a degree of utility for using WhatsApp for qualitative data collection. The app’s multifaceted nature, cost-effectiveness, convenience and ease of use for both researchers and participants is undeniable. These advantages make WhatsApp a favourable tool for data collection, especially in African countries like South Africa, affording participants the opportunity to participate in qualitative research since they do not have to take time off from work. This allows more kinds of people to contribute to research studies such as this one on the drivers of information disorders in the Global South.
