What is Social Desirability Bias?
Social desirability bias is the tendency for people to answer questions in a way that will be viewed favorably by others, rather than honestly. It can make respondents over-report 'good' behaviours and under-report 'embarrassing' or unpopular ones.
Social desirability bias occurs when survey respondents shape their answers to appear more socially acceptable, virtuous or normative. This happens for sensitive topics (e.g., voting, substance use, charitable giving, discrimination) and when respondents think their answers could be judged. The result is systematically biased data: behaviour or attitudes appear more socially approved than they really are. It can be influenced by question wording, survey mode (phone vs. anonymous web), cultural norms, and — in multilingual surveys — translation choices that sound more or less face-saving in different languages.
Usage example
If you ask “Do you always recycle?” many people may answer “Yes” even if they do not, because recycling is socially approved. Rephrasing to “How often do you recycle?” with frequency options and an anonymous response mode reduces the pressure to give the 'right' answer.
Practical application
Why it matters: social desirability bias reduces the accuracy of survey findings, especially on sensitive topics, and can lead organisations to make wrong decisions about services, outreach and policy. For Hearo users this has extra importance because: different language communities may have different norms about politeness or face-saving, and automatic translations can unintentionally increase pressure to give socially acceptable answers. Practical steps to reduce it include: - Make surveys self-administered and clearly anonymous (no participant account required). - Use neutral, non-judgmental wording and offer balanced response options (e.g., frequency scales rather than yes/no). - Provide a “prefer not to answer” option for sensitive questions. - Pilot questions and review translations with community members to avoid phrasing that sounds leading or shaming. - Where possible, use objective or behavioural questions (e.g., “How many times in the last week…”) or triangulate with administrative data. - Place sensitive questions later in the survey after rapport is established, and remind participants of confidentiality. These steps help collect more reliable responses from multilingual audiences and reduce the chance that some communities are misrepresented.
FAQ
How is social desirability bias different from other survey biases like acquiescence?
They’re related but distinct. Acquiescence bias is the tendency to agree with statements regardless of content; social desirability bias is deliberately giving answers that seem socially acceptable. Both can coexist and should be addressed with careful question design, balanced wording and varied response formats.
Will making the survey anonymous eliminate social desirability bias?
Anonymity reduces the pressure to give socially acceptable answers but doesn’t eliminate the bias entirely. People may still self-present positively out of habit or internalised norms. Combining anonymity with neutral wording, indirect questions, and response options that reduce judgement works best.
Can translations make social desirability bias worse?
Yes. Some phrasings in another language may sound more moralising or face-threatening, increasing pressure to give the 'correct' answer. That’s why translation review, community testing, and hearing participant feedback on wording are important parts of multilingual survey design.