Ethical considerations in conflict and development research

"An essential ethical requirement for any survey is, for example, that participation is voluntary, uncoerced and can be terminated at any time."

The first "gene-edited" babies in China, the fast-track approval of novel vaccines during the COVID-19 pandemic, and currently the general availability of generative language models such as ChatGPT: When matters of scientific ethics are discussed by the general public, they tend to revolve around medical ethics, disruptive technologies developed in the computer and natural sciences, or animal experiments in research.

In many of these areas, clear legal provisions regulate at a state, federal or EU level what is allowed in research (and what is not). However, what is the situation in research areas where the rules are less strict? Which authorities oversee and regulate such work, and how much responsibility lies with the researchers themselves? We interviewed Anke Höffler from the Department of Politics and Public Administration to find out. The economist and Alexander von Humboldt professor investigates the social causes of violence in developing countries and conflict-ridden regions, among other things.

Ms Höffler, when do ethical considerations play a role in your research projects?

That depends primarily on the specific research question and the methodology required. There are projects in my research area that are mostly unproblematic – for example, when we analyze publicly available macroeconomic data from the World Bank. The most pressing question there might be how the findings could be used for political purposes. There are, however, more critical research questions that we can only answer by interviewing people. This is especially true with topics related to violence, as is the case in my research. In addition to considerations regarding data protection and the subsequent use of the research findings, there must be a strong focus on safety and ensuring that no one gets harmed: neither the participants nor my staff.

Can you give an example to illustrate how you would need to protect your staff?

We are currently working on a study on patterns of sexual violence among young people in Nigeria. To do this, we asked young people between the ages of 13 to 17 – including those who are not currently in the school system – about their experiences with violence. This, of course, is only possible with the support of local partners. They know the local situation and rules of conduct and have an easier time connecting with residents. Nevertheless, very sensitive situations can arise during such surveys. In the case of this particular study, for example, we decided to conduct part of our survey using pen and paper. This would normally not be the method of choice, because the data then need to be entered by hand so they are available and can be evaluated digitally. However, we did not want our staff to take laptops or tablets with them as this would have exposed them to an unnecessary risk of robbery.

So, in your ethical considerations, is it a question of weighing the various risks?

Often yes. In this case, we accepted the relatively low risk of isolated errors when transferring the data in exchange for being able to extend our survey to young people outside the school system without exposing our on-site staff to increased danger. Of course, the safety and rights of the interviewees are just as important as the safety of our staff. For example, an essential ethical requirement for any survey is that participation is voluntary, uncoerced and can be terminated at any time. This also means that respondents must be informed about the purpose of the survey before they agree to participate – so they can give informed consent.

At first glance, this sounds trivial, but what are some of the problems that can arise?

"In our study in Nigeria, for example, many of the young people we interviewed were illiterate. This required very clear information in very simple language, which we played to them via audio recording. Some of the participants gave their consent by fingerprint instead of signature. The survey itself, of course, must also be as easy to understand as possible and written in the local language – which, in the case of the Nigeria study, is Yoruba. All of this is very time-consuming and complex. For this reason, every new survey study is preceded by a pilot study to ensure that everything runs smoothly before we roll out the full study. This is extremely important because you almost always discover weaknesses that were overlooked in the design phase."

Anke Höffler

I can give a very distressing example from one of our studies on family violence in Kenya. We had asked mothers how often they beat their children with a cane. After the pilot study, we had to add the answer "multiple times a day" to our response options. Not even our local partners had considered this as an option when designing the survey.

You said earlier that the safety of the interviewees is also an issue. To what extent can their safety be compromised?

There are several levels to consider: Interpersonal violence, for example, very often takes place in the immediate environment of the victims. We must therefore ensure that the act of sharing their experiences of violence with us does not make our participants targets yet again. Imagine a violent husband who realizes that his wife has told us about her traumatic experiences and who punishes her with more violence as a result. We must take utmost care to prevent such situations, for example by conducting the interviews outside the home. In cases where this is not possible, we sometimes give the respondents neutral questionnaires that conceal the actual topic of the survey. They can then show these to others around them in case they have any questions. As you can see, ethical considerations for researchers often involve methodological details that urgently need to be clarified in advance.

Is all this your personal responsibility?

In the initial planning stage, yes. We first discuss new research projects as a team. I am fortunate to have a very interdisciplinary group that includes economists, political scientists, psychologists, among others. This allows us to draw on a wealth of experience and consider multiple perspectives. This is complemented by the expertise of our local partners, who also draw our attention to local particularities. This exchange among colleagues is extremely important when planning projects and preparing ethics applications. We often seek additional advice internally from the Research Support Unit at the University of Konstanz, or – depending on the research question – externally from organizations that have experience with certain sensitive topics and can provide expert advice. For a study on domestic violence, for example, we consulted with UN Women. Various committees are involved in deciding whether a study can actually be carried out as planned, based on our proposals.

Which committees are these usually in your case?

At university level, the Ethics Committee reviews research projects involving human test subjects. We would not launch a study without their positive vote. Our international partners also have similar institutional committees. Different universities or countries have different ethical requirements. Of course, we also have to take into account the guidelines for all the parties involved. Finally yet importantly, when conducting our studies abroad, we usually need official approval at the state or at least district level, as well as from the management of the respective institutions, for example, if we work with schools or hospitals. The willingness to cooperate in this regard is generally very high though, because the countries we work in often have legal requirements – for example, for implementing the UN Convention on the Rights of the Child. And this is something our research helps them with, as we compile current information on relevant topics and share our findings with the corresponding institutions and politicians.

Are there any ethical aspects that need to be considered after a study has been completed?

Yes, there are quite a few. These range from data protection issues – for example, the anonymity of respondents must be guaranteed at all times – to the potential misuse of research results, and the question of how to make the study results available to as many people as possible in a transparent way in the spirit of open science.  There is, however, another point concerning the people involved in the study that is very important to me and which I would like to mention in conclusion.

"The people who take part in our research have often suffered severe violence. We therefore have to take care when designing the surveys so that the respondents are not re-traumatized by having to recall their experiences. If respondents need them, appropriate support offers must be in place."

Anke Höffler

Are there indicators of this happening frequently?

Fortunately, research on this topic has found very little evidence of high levels of re-traumatization among survey participants. In fact, we often get feedback from our respondents that they were happy to be able to speak out – that their voices and experiences are being heard. Nevertheless, in our projects, we make sure that we can intervene on the spot in the event of an emergency, and offer our study participants advice or point towards contacts that can help. Our study leader in Nigeria, for example, is a psychologist and is available to the young participants as a point of contact after the survey is finished. Not in the role of a therapist, but as someone who can put them in touch with legitimate and trustworthy help. The same applies to our own staff, of course, who have to deal with distressing topics for long periods of time when conducting these studies. This is why we are currently in the process of organizing psychological supervision for our staff, as there is no such service available at the university level yet. It is not only physical safety that we need to ensure for everyone involved in our studies, but also psychological safety.

In 2018, economist Anke Höffler was appointed to an Alexander von Humboldt Professorship (Development Research) in the Department of Politics and Public Administration at the University of Konstanz. Her research focuses on the social causes of interpersonal and collective violence, among other things. She is a principal investigator in the Cluster of Excellence "The Politics of Inequality".

Header image: Universität Konstanz / Inka Reiter.

Daniel Schmidtke

By Daniel Schmidtke - 12.01.2024