First time doing this so apologies if I mess up.
We ran into a situation where we were experiencing a lot of respondent fraud. Either that, or the entire faculty of a large academic medical center in New York City moved en mass to Dubai. It was clear they were rubbish from their other responses, but it got me to thinking whether we could have used earlier screening questions to avoid them getting through the survey.
We asked respondents to identify the state where they were licensed to practice medicine. Is there a way to validate that if the long/lat coordinates are clearly "not" in New York or at least the north east, then this would be flagged for termination?
Interestingly, all the people who located near their stated state of practice seemed to give intelligent answers to questions, where as those who did not were clearly garbage answers.
I thought authenticator might help, but it doesn't seem it will. I don't want to locate my respondents precisely at all. I just want to know whether your metadata puts you anywhere close to where you say you are.
Any thoughts would be appreciated.
Best answer by AxelS
View original
