The challenges with data confidentiality in developing countries

Illustration: TBS

Data collection in developing countries – specifically in social sciences – comes with unique challenges, including unreliability of census data, enumerator misconduct, handling topics that are socially, culturally, and politically sensitive, and respondent unease with the interview format, to name a few. 

These challenges are well-documented in the literature.

There is, however, another issue that has received less attention in literature. Ensuring participant protection through data security and confidentiality is often severely lacking and has not been discussed much. International donors and grant organisations often make demands contrary to the spirit of confidentiality in developing countries.

Cultural, social and political mores differ greatly enough to require different strategies from the standard practices of the developed countries. In many developing countries, not necessarily only the non-democratic ones, answers to questions that carry with them political or social connotations may jeopardise the safety, livelihood and social standing of respondents if made public.

The World Bank, one of the biggest funders of surveys and research into socio-economic factors in the developing world, considers this one of the top eight priorities for household surveys for the next decade: improve data access, discoverability and dissemination.

In social sciences especially, data collection strategies must be tailored carefully to fit the unique local characteristics of research sites, with a greater focus on data security and confidentiality. Unfortunately, evidence suggests that the standards of data confidentiality are lower in these regions. There are various reasons for this.

The Institutional Review Boards (IRB) 

Survey works involving human subjects often require approvals from IRBs. Most IRBs in developing countries may focus primarily on clinical research and are unfamiliar with assessing social science research. 

For example, the IRB may need all survey participants to provide written consent. However, written consent also increases the risk of participants' identities coming out. 

In addition, IRBs may require that surveys be conducted with higher 'visibility' than necessary or require a painstakingly tedious and thorough process of administering questionnaires to be followed. 

All these – if not modified suitably to fit the unique characteristics of the research site – jeopardise research validity (participants themselves may be unwilling to speak honestly if they do not feel safe to do so) and increase the risk of harming confidentiality.

Researchers and enumerators 

Going beyond IRBs, the researchers often implement rules that increase the risk of compromising data confidentiality. The issue of data fabrication in the developing world often makes the researchers require the survey enumerators (often grad students, interns, or hired third parties) to provide evidence that data is being collected properly. This may include audio recordings, pictures of interviews or group discussions, and the collection of contact information (for example, phone numbers and home addresses).

Since I started conducting surveys in 2013, I have never conducted survey work in rural Bangladesh, where the PI (and the donor) did not insist on pictures and contact information of the participants. Be it a consulting project funded by a leading multilateral donor agency or a grant project funded by a public agency of a foreign government. It is unlikely that the same institutions would make similar demands if I were doing the same work in the United States, for example.

Across the developing world, these are standard practices. A survey participant who revealed compromising information about their preference of political parties, for example, may have their picture taken, name and home address and phone number jotted down on the questionnaire, which is then uploaded to a for-pay server or a researcher's work-computer, where it stays for years. The hardcopy is thrown into the enumerator's backpack, handed over to the researcher at the end of the survey, where it sits in an unsecured storeroom for years.

Gaps in digital literacy

Even for seasoned researchers, understanding what constitutes safe data storage and what is risky is sometimes lacking. Sensitive data is copy-pasted onto an office hard drive or a for-pay storage such as Dropbox or OneDrive. 

Data is shared as email attachments (in email threads that run into tens of replies and forwards to numerous parties) and carried around in external storage devices such as flash drives. 

Cost constraints (stricter in developing countries) may mean secure servers are unavailable; researchers may not know data storage best practices, etc.

Voice of the marginalised 

The risk to a researcher due to data leakage leading to harm from a survey participant is low to non-existent if the victim is marginalised. There may not be any competent advocacy group to take up their cause, there may be no legislative redressal mechanism, and the victims themselves may not know or adequately understand they have been 'wronged' and owed something. 

Such a reality makes researchers, donors, and funding agencies prioritise research outcomes above all else.

The World Bank, one of the biggest funders of surveys and research into socio-economic factors in the developing world, considers this one of the top eight priorities for household surveys for the next decade: improve data access, discoverability and dissemination. 

The document mentions the need for greater use of digitisation, AI, para data and metadata, etc. Nowhere in the document do they mention the need for greater protection for survey participants or survey data.

It would be unreasonable, perhaps, to expect local authorities to take up the cause of protection of survey participants when their resources are already stretched so much, and their priorities lay elsewhere. The onus, therefore, should be on the institutions – especially the international donor agencies and grant organisations – and the researchers to ensure that better strategies that account for the ground-level reality of developing countries are devised and implemented.

Ahsan Senan. Illustration: TBS

Ahsan Senan is an economist.

Disclaimer: The views and opinions expressed in this article are those of the author and do not necessarily reflect the opinions and views of The Business Standard.