Social desirability bias is a type of response bias in which people tend to answer questions according to how their answers will be viewed by others instead of answering truthfully. Respondents may give you answers that are more favorably acceptable or more in line with the popular opinion or politically correct response. The answers they give may be inflated to reflect “good behavior” or under-inflated to hide “bad behavior”.
Topics that are sensitive to social desirability bias include self-reported personality traits, socioeconomic status, religion, charitable acts, personal habits around smoking, drinking, drugs and gambling, illegal behavior and other controversial subject matters. For example, if a survey included a question “How many alcoholic drinks do you consume in a day?”, the respondent might answer 1 or 2 which seems to be socially acceptable rather than the 4 or 5, which they truly consume.
Tips to Combat Social Desirability Bias in Surveys
- Use a social desirability scale. This was devised by Crowne and Marlowe to gain insight into whether people tend to give socially desirable responses. It involves asking a series of questions used to identify where one falls along the scale. Based on respondents’ scores, you can then determine how many respondents with particular scores to include or exclude from your survey.
- Validate with other data. For socioeconomic status questions, you can cross reference other self-reported information in the survey. If someone says they earn over $150K in a single person household, but also said they are a public elementary school teacher and are only age 24, it is likely that the income reported is over-inflated and you may choose to remove that respondent’s data from your survey.
- Communicate and implement anonymous random model surveys. If a respondent knows his answers won’t be shared with anyone and personal details won’t be disclosed, he will be more likely to answer truthfully. Begin your survey with an introduction telling respondents the answers they give will be kept confidential.
- Keep the purpose of the survey vague. If the main purpose of the survey is known or the company conducting the survey, respondents will be able to prepare and give responses that are more socially acceptable. For example, if a survey is sent from an organization that supports animal rights (such as PETA), people may underestimate the amount of meat they eat or express an overly negative attitude toward buying fur coats.
- Wordsmith your questions carefully. You can provide statements to the respondent and mention that these were statements made by other people. Then have them choose the one they relate with the most. This way they know any response is perfectly acceptable and they don’t have to worry about writing their own. Another technique is to reinforce before or after the question that there are no right or wrong answers.
Social desirability bias prevents people from giving truthful answers to survey questions, leading to skewed results. The entire purpose of conducting surveys is to obtain information that is based on respondents providing honest answers. Therefore, it is important for you to be aware of social desirability bias and design your survey to minimize this bias as much as possible.