Data and International Development: Insights From Online Information Seeking on the Coronavirus in Nigeria

By Babatunde Okunoye

In developing countries, where the development need is greatest, the data ecosystem to enable and inform development is severely lacking. Developing countries sometimes lack robust and comprehensive data which is the basis for development planning. Within these contexts, it is often the case that public data collection for national accounts, household and firm surveys, data collection through administrative systems such as birth records, pensions, tax records, health and census are performed infrequently and often lack the granularity necessary to make meaningful inferences about small, sub-populations of interest. Data is scarce where it’s most needed.

Nevertheless in the past decade, within these same developmental contexts, there has been a massive proliferation of mobile devices and access to the Internet. Access to these devices have inspired new and refined BigData research approaches in the service of development. In contrast to some early scholarship on BigData for development research which lacked scientific transparency, displayed methodological weaknesses and ignored the continued relevance of ‘’small data’’ traditional research methods such as surveys and qualitative interviews; recent work by numerous organizations around the world and embodied by the World Bank’s Mind, Behaviour and Development unit have spearheaded a new approach to BigData development research which adequately blends the strengths of BigData and traditional research methods for development.

Talking about development, the ongoing coronavirus pandemic has laid bare the deep development chasm which exists in the world. Although much of the disease burden is in countries of the Global North, countries of the Global South might be worse off after the pandemic because of the state of their economies, health infrastructure, and information systems.

For instance, many developed nations have already implemented some forms of financial support for their citizens to help them cope with the income loss resulting from national lockdown orders. Developed nations of the Global North also have some of the most advanced health systems in the world, although many are now strained due to the burden of the pandemic. These nations also tend to have varied and reliable public and private information sources where citizens can draw accurate and timely information on the status of the pandemic and other helpful tips to stay healthy.

The situation described above cannot be said to be true for some of the world’s developing and underdeveloped countries. Many have ordered national lockdowns without insufficient financial support to citizens and businesses. Developing nations also have some of the weakest public health systems in the world. In these development contexts also, information sources are not always as varied and reliable as sources in more developed country contexts.

In these constrained information ecosystems present in developing nations, a key component of information seeking by citizens will be online sources accessed by search engine queries. The use of search engines is one of the most popular uses of the Internet. Search engine use is the most popular approach to online information seeking, and almost half of all Internet users now use search engines on a typical day. They are for many people the first page they see online, thus serving as the gateway to the Internet. Search engine queries when aggregated and analysed using BigData methods can become a useful window to understanding developmental needs of society or at least segments of society.

As a case study, I analyzed data trends for online searches on the search terms ‘’coronavirus’’ and ‘’covid-19’’ on Google for the past 90 days in Nigeria (Figures 1 and 2). This represents a time span from the beginning of the year to the last week of April, and corresponds to the spread of the virus globally. Google is the most widely used search engine in Nigeria, and the world.

Figure 1: Search trends for the terms ‘’coronavirus’’ and ‘’covid-19’’ in Nigeria for the past 90 days, accessed April 28 2020.

Figure 2: Regional search data for the terms ‘’coronavirus’’ and ‘’covid-19’’ in Nigeria for the past 90 days, accessed April 28 2020.

A cursory inspection of the search trends reveals that search interest in these two terms in Nigeria roughly corresponds to global search trends (Figures 3 and 4). These trends correspond to the spread of the pandemic globally and locally in Nigeria, with minimal interest during the first weeks of the year with a spike in March as disease cases peaked and taper off off in April as disease curves began to flatten and national lockdowns began to be eased, including in Nigeria.

Although the similarities of online search trends on the coronavirus with the global and local spread of the disease (in Nigeria) is interesting, delving into the local data on Nigeria reveals interesting facets of the data relevant for development. Global search trends of ‘’coronavirus’’ show that the most frequent searches for the term were from Italy, Spain, France, United Kingdom and Switzerland (Figure 3). This data is not surprising given that Italy had been the epicentre of the pandemic in Europe, and Spain, France, the United Kingdom and Switzerland are all European countries at the heart of virus transmission.

Figure 3: Search trends for the terms ‘’coronavirus’’ and ‘’covid-19’’ globally for the past 90 days, accessed April 28 2020.

Figure 4: Regional search data for the terms ‘’coronavirus’’ and ‘’covid-19’’ globally for the past 90 days, accessed April 28 2020.

The data on Nigeria is however startling. Three of the top five states in Nigeria (with 36 states) with the most frequent searches for coronavirus are regions within Nigeria’s north which has experienced lack of economic investment and has the highest poverty rates, highest illiteracy rates, least development and minimal 3G telecommunications access (Internet access in Nigeria is largely carried by mobile networks). Furthermore, two of these states — Borno and Yobe have been at the heart of the violent insurgency of the past decade with the destruction of communities and infrastructure. Their topping the search frequency represents an outlier to say the least.

Within the online search data for the coronavirus pandemic related information in Nigeria, there are developmental insights to be mined. Some might include ‘’how is it possible that these states with developmental challenges described above have the most frequent searches for coronavirus-related information?’’ or ‘’What interventions can be implemented to support these information ecosystems in the midst of this pandemic and beyond?’’ Although these overall trends were brought to our attention by big data such as aggregate search engine queries, making sense of what’s happening locally will require traditional ‘’small data’’ research methods such as local representative surveys, qualitative interviews, and focus group studies. As has been rehearsed earlier, only by combining data from traditional social science methods and emerging big data methods can we arrive at a deeper and clearer understanding of our world.

Babatunde Okunoye is a researcher on digital society. He led research at Paradigm Initiative, a digital rights and inclusion organization in Africa. Babatunde’s research focuses on the use of aggregate search engine queries to inform public policy and international development, particularly in statistically poor contexts of developing countries