Users tend to skew survey answers towards what's socially acceptable.
<aside> ℹ️ Survey Bias occurs when people ruin the survey results by not answering truthfully but rather than choosing what’s socially acceptable.
</aside>
Social desirability bias is when people tend to give answers that they think are socially acceptable or desirable rather than providing their true beliefs or behaviors. This trend is expressed in an exaggeration of positive and desirable qualities and behavior and an understatement of negative, undesirable.
For example, if you are asked how often you exercise, you may be tempted to say that you exercise every day, even if you only exercise a few times a week. Social desirability bias can impact the results of a survey by causing people to over-report positive behaviors and under-report negative behaviors.
In fact, social desirability bias can work both ways. Depending on the question you can find both over-reporting and under-reporting of a particular viewpoint, and that can have a dramatic affect on the usability of the data eventually generated.
Understanding the effect of social desirability will allow us to reduce the noise in the data obtained from such studies. So, to reduce the impact of this bias in polls, we can add anonymity to the participants.
In questionnaires, we can play with the semantics of the questions in a way that makes it more comfortable for users to answer openly. In an interview, we can introduce the interviewer to the users in advance, perhaps even discuss a few informal questions, and later leave the interviewer alone with the user.
When it comes to app functionality, we need to understand that we will never be able to keep our users on the project, no matter how cool software we have written for them, if using it means for them going against social norms.
Socially Desirable Response Tendencies in Survey Research