There are numerous methods of estimating errors in public opinion research: it's a problem that some people lie on surveys, either because they don't think it is safe to answer or because they...
There are numerous methods of estimating errors in public opinion research: it's a problem that some people lie on surveys, either because they don't think it is safe to answer or because they think it is amusing to lie. The first problem can be addressed by asking "list" questions where you ask people to count how many things they've done from a list that includes safe and non-safe items and showing different people different lists. The second problem is addressed in this study by adding questions that invite invalid answers and assuming that those people's responses aren't trustworthy.
They find that the answers don't change much for responses that are relatively frequent such as "did you consider suicide?" but it does make a difference for less common responses like "do you take steroids?" as wel as surveys on strange behaviors like
There are numerous methods of estimating errors in public opinion research: it's a problem that some people lie on surveys, either because they don't think it is safe to answer or because they think it is amusing to lie. The first problem can be addressed by asking "list" questions where you ask people to count how many things they've done from a list that includes safe and non-safe items and showing different people different lists. The second problem is addressed in this study by adding questions that invite invalid answers and assuming that those people's responses aren't trustworthy.
They find that the answers don't change much for responses that are relatively frequent such as "did you consider suicide?" but it does make a difference for less common responses like "do you take steroids?" as wel as surveys on strange behaviors like
https://www.sciencedirect.com/science/article/abs/pii/S1471015323000053?via%3Dihub