Perception can be measured by adopting models used in the social sciences. This study tried to understand how MSF’s image is constructed and how it is conveyed to audiences outside the organization.
Similarly, it is important to find out how this institutional identity is received and understood by the people who interact with MSF as an organization, employer, and medical structure.
In addition to the issue of organizational image, certain other questions needed to be explored:
• How is humanitarian aid—as opposed to development assistance—generally perceived by the host societies? It is obvious, though not always remembered, that those to whom humanitarian aid is delivered are not simply passive recipients.
This investigation may appear biased, given that MSF is investigating itself. We decided at the outset that the aim of the project was not to carry out an exhaustive analysis, but rather to improve the way projects are implemented and perceived in the field. Therefore, we accepted the possible bias this situation might produce.
The whole process of the study has been extremely valuable. Indeed, it prompted wide debate within the teams, particularly those working on projects, in capital bases, in the “cells,” and throughout MSF-Switzerland and the various departments. Trying to learn how others perceive MSF finally led the organization to try to define itself and gain a better understanding of the evolution that has taken place in its human composition, its actions, discourse, and so on.
A multifaceted methodology was adopted in order to refine notions of perception, prepare ways of comparing one context to another, and, in particular, to issue practical recommendations to enhance the organization’s medical activities.
Choice of Projects Visited in the Field
At the start of the study, we envisaged the need to explore three types of contexts:
• Contexts of conflict/tension where, for security reasons, the organization must distinguish itself from other foreign (humanitarian, military, political, etc.) actors, such as in Darfur, the Democratic Republic of Congo (DRC), Chad, and Iraqi Kurdistan;
By conducting numerous field surveys, it would be possible to validate the results and identify trends in perceptions of humanitarian action and MSF. One of the key hypotheses was the importance of issues of perception, particularly in contexts of conflict or tensions.
To avoid being overly influenced by the Operations Department, we decided in 2007 to divide the projects between stable and unstable contexts. Within these two groups, we made a random selection of six projects. This distinction (stable vs. unstable contexts) is itself extremely debatable and corresponds more to a wish to distinguish between projects where problems of perception might give rise to the need for security measures for the teams and projects where perception was more to do with achieving optimum implementation of the medical project.
Of the 12 projects selected at the start of the study (DRC, Darfur, Iraq, Somalia, Chad, and Myanmar for so-called “unstable” contexts, and Cameroon, Kyrgyzstan, Kenya, Guatemala, Liberia, and Niger for so-called “stable” contexts), not all were visited, for two main reasons. The first was security problems (the bombing of the Seleia camp in Darfur in 2008 took place one week before the planned visit which, for obvious reasons, was canceled). The second reason was the relative similarity of the results obtained in previous surveys, beyond the specificities of each context. It seemed pointless to spend too long carrying out field visits. Consequently, the central aim of this research quickly became the appropriation of the results by all the teams.
A pilot study was conducted in September and October 2007 in Zinder, Niger. Following that visit some questions were tweaked, but on the whole the methodology was considered satisfactory.
A number of key methodological techniques were used:
Preliminary Literature Reviews
All the literature on each MSF project was analyzed, bearing in mind the context of the country of intervention. The aim of this stage was to gain an understanding of the environment in which the project was implemented: who makes the decision to intervene (MSF or the authorities of the country in question), the context of intervention (acute crisis, conflict, nutritional crisis, stability, etc.), the history of humanitarian action in the country, and analysis of tensions with the population and/or local authorities.
A questionnaire was prepared. It was quite long and grouped questions around several themes: the perception of humanitarian action (Where does it come from? Whom does it serve? Who provides it? What is its impact? What criteria are used to measure the quality of aid?); the perception of MSF (How are the principles upheld by the organization understood? How does it set up its projects? Is it considered transparent? What is its origin? etc.). For vertical programs, what is the perception of the disease treated? This last question was added following the first field visit, as we realized the importance of this factor in how the organization is perceived. Later on, we will look again at the modifications made to the questionnaire over the course of the project.
The questionnaire was translated into all the necessary vernacular languages: French, English, Hausa, Arabic, Spanish, Russian, Pokot, and Liberian English. Inevitably, some nuances were lost in translation, but every effort was made to minimize those losses. Indeed, a key challenge was the process of defining certain terms, such as the principles of impartiality, neutrality, and independence.
We chose to work in collaboration with local universities. Thanks to the heads of the departments concerned, several master’s degree students (of sociology, anthropology, and political science) were selected and trained in the questionnaire and in leading discussion groups. The aim of the training was to ensure that all the questions asked were understood and made sense in the contexts in which we work, and that the translation from the source language into the target language was correct. It was also an opportunity to work with the students on techniques for interviews in small groups and semi-structured interviews with several people. It is interesting to note that this research aroused a great deal of interest within the faculties themselves, as it was an opportunity for students to reflect on the ins and outs of the humanitarian aid delivered in their own countries.
The students led discussion groups based on the questionnaire. These groups consisted of 10 to 15 people and an average of 50 groups were held for each visit. So, more than 600 people were questioned at each field site. From the pilot visit, it emerged that it was preferable to organize groups according to categories. Consequently, we gathered the opinions of national staff; international staff; people living near MSF facilities; people not necessarily in daily contact with the “MSF apparatus;” patients and their families/companions; local authorities (e.g., the Sultan in Zinder, Niger); administrative, religious, and political authorities; those responsible for health at the local, regional, and national levels (e.g., the Ministry of Health); traditional practitioners or local doctors; UN agencies; local associations; local or international NGOs; other MSF sections; other international actors (e.g., the Humanitarian Aid Office of the European Commission); and armed groups present in the region. This list is not exhaustive, but gives an idea of the range of stakeholders that we contacted within the framework of this study.
Once these groups had been identified, we arranged interviews. The discussion groups were set up based on two criteria: their relationship with MSF (beneficiaries, those close to the project, those neither beneficiaries nor close to the project) and their sociological component (sex, age, and role in the population).
It is interesting to note that, at the start of the field visits, those most reluctant to take part in the study were the international staff, who perceived it as an assessment of their own work. In contrast, the national staff were more receptive, as it gave them an opportunity to voice their wishes with regard to the management of human resources and gain better access to information about the organization.
Following the discussion groups, semi-structured interviews were conducted with a number of stakeholders considered more “sensitive”: some local and national political authorities, religious authorities, armed groups, etc. The same guide was used for all the semi-structured interviews. The whole structure was reproduced identically at every field site in order to enable comparison. A series of semi-structured interviews were carried out with specific stakeholders: the leader of the Zakat committee and local authorities (mayors, district chiefs, etc.) whose positions made a group discussion difficult.
The people interviewed were not asked to sign a consent form, as they are not identified by name in the reports.
A scientific committee was set up to monitor the project. Throughout the study, this committee offered advice on refocusing the research and responding to the different reports produced. In September 2008, after a few field visits had been completed, a number of methodological and practical modifications were made following the discussion of the project by the scientific committee. The idea was to change from a semi-quantitative approach to a much more qualitative approach, using discussion groups as a way of exploring the perceptions of the people questioned in greater depth. These adaptations were possible thanks to the field visits already carried out and the identification of certain recurrent themes across the different sites, and concerned the projects in Guatemala, Kyrgyzstan, Cameroon, Iraq, Jordan, and the Occupied Palestinian Territories. The idea was to try to deconstruct the discourses of all the actors and extract the discursive reasoning behind them. This made it possible to analyze the potential role of national staff as “development brokers.” Linear interviews were abandoned in favor of an exploration of themes and possible logical links around those themes. Depending on the field site and the needs of the Operations Department, a series of specific questions about concerns in the field were also introduced.
Following each field visit, a report was drafted. The target audience of these reports consisted of the field teams and the teams managing the projects at the headquarters. They contained both an analysis of the project visited and practical recommendations, some of which were implemented quite soon after the survey.
Other Research Into Perception
Another task consisted of systematically reviewing what had already been written about perception in recent years. This research project was inspired by several other projects carried out by various other institutions and greatly benefited from the involvement of external participants. Béatrice Pouligny wrote
At the same time, MSF Belgium conducted a survey of perceptions in the DRC (questionnaires and focus groups) and Rwanda, while MSF Holland studied Haiti. These surveys were mainly prompted by tensions within the MSF teams, however, to gather the opinions of national staff members. It is important to underline that the research performed by MSF-Switzerland differs from these previous studies in two ways– first, it has an operational aim and, second, it covers a much broader field of investigation than the earlier work.
The process of linking up with other research bodies or specialists working on these themes for other aid organizations made it possible not only to better define the subject and clarify areas where a more in-depth examination was necessary, but also to develop the most appropriate methodology for implementing the survey at the chosen field sites.
It was believed that an MSF project site would be the best unit of measurement, as the effect of the MSF mission is visible through its action in the field, the scope of which is mainly local. It was likely that within a single context, local people’s perceptions of MSF might differ from one project location to another.