Language Technology to Improve Data Quality and Engagement

Needs assessments are inaccurate and slow in multilingual contexts, a poor basis for humanitarian strategies. TWB’s research shows that data collectors often do not speak the languages of affected people, do not understand humanitarian terminology and have limited training. Information is translated 2 or 3 times, making the process slow and error prone. The voices and needs of affected people are lost; aid efforts are designed and continued based on incomplete or inaccurate information.