Table 4 Case example: the importance of appropriate terminology and accurate translation

Countries: Bangladesh, Myanmar, DRC

Sector: General

During the Rohingya response, practitioners reported struggling to translate program information and interview questions into both Burmese and Rohingya, especially because Rohingya is not a written language. Determining direct translations for words like “gender” or “safety” was a challenge. As a result, one organization in Myanmar used a literal translation of “gender-based violence,” which translated as “sex-based violence.” This translation led to confusion among the organization’s staff, some of whom misinterpreted this to believe that physical violence was not a form of gender-based violence. A different organization used a longer translation that included all of the different types of gender-based violence for clarity. This definition seemed less efficient due to its length, but it yielded more accurate information

Similar issues were described by a report on measuring gender-based violence in peacebuilding contexts [18]. When one organization tried to measure the level of female participation in communities in North and South Kivu provinces in eastern DRC, they struggled to translate between local languages and Swahili. Questions about women’s participation in community meetings were often understood to simply mean women’s presence, which did not fully capture the level of engagement that researchers were trying to measure. Taking these translation and interpretation issues into account, the organization expanded the indicator to include voicing opinions as well

Key take aways: Work with skilled translators and cultural mediators to translate M&E tools so that both local humanitarian staff and community members fully understand the terms and questions. Simpler terms may be more appropriate. Use back translation—a process where a different person translates the question that has been translated into the local language back into the original language—to understand where meanings may be lost or distorted. Always pilot questions with individuals representative of the affected populations before broadly implementing questions