Translation In Multinational Polls Could Cause Bias

by Stratton Craig

image overlay

The advancement of modern technology has allowed many businesses to go global and this is undoubtedly a great advantage for international brands which can rely on broad audiences and large markets. However, brands should be aware that when they carry out multinational research, results may be influenced by the translation of the terms used in the surveys, leading to bias in responses.
This is what a recent study published in the Journal of Consumer Research suggests, highlighting the importance of proper translation and taking cultural differences into consideration.
Researchers found that consumers were influenced in their answers by the labels used to mark the endpoints of a survey. If these labels are not fully equivalent in multilingual research and at the same time sound unfamiliar to the respective groups of respondents, this could affect results, they explained.
Many consumer polls rely on response scales that measure respondents´ agreement or disagreement with given statements, with some the most commonly used categories labelled as “strongly (dis)agree” or “completely (dis)agree.” Researchers asked English and French speaking consumers in the UK, US, Canada, and France to take part in a similar survey and found that responses were influenced by the familiarity with labels used for the different categories. Both English-speaking and French-speaking consumers tended to choose the label that was more frequently used in their everyday language – completely agree or tout à fait d´accord. By contrast, when the label said “extremely agree or extrêmement d´accord”, consumers were significantly less likely to select them.