What percentage of your income do you give to charity? If you saw an injured bunny rabbit, would you try to help it? These are questions with a "right" answer. And there's another survey to determine whether the person answering is likely to give that "right" answer, no matter what they'd actually do.
People lie to surveyors. They always have and they always will. This is why people adjust the number of sexual partners they'll claim if they're made to believe that they're hooked up to a lie detector. It's why kids will claim they've done drugs that don't exist. There is simply no amount of anonymity that an interviewer can give, no level of assurance that they've heard everything before, no guarantee that they just have a dispassionate interest in the facts, which will completely stop people from lying. And that's when the interviewer is telling the truth about how anonymous the survey is, and how much they don't personally care what the answer is.
When people are giving answers that could potentially damage their futures, they'll lie their heads off. Sometimes they won't even know they're doing it. We naturally tend to search for our best moments and exaggerate our best qualities when we're thinking about ourselves. How does an interviewer control for this need to display socially desirable qualities?
One way is to give people surveys that measure how much they'll overemphasize their own good qualities. These tests often consist of nothing more than a long list of qualities that everyone is supposed to have, like "I always try to practice what I preach." Others incorporate obviously undesirable qualities. How many characteristics the respondent checks off, and how many bad characteristics they'll deny, will usually indicate how much they would like to report about themselves. The most famous, and most used, is the Marlowe-Crowne Social Desirability Scale, which has been used to suss out the self-esteem levels of everyone from executives to sex offenders. You can go ahead and take it yourself at this link.