There's no way that anybody studying any portion of the sex industry can guarantee the truth of any account, whether it be a sex worker, sex buyer, manager, owner, operator, policeman, lawyer, or politician. What we can do is construct our instruments in such a way that we phrase questions that are not leading and that do not result in bias, that do not result in response bias, in social desirability bias. This is much easier in self-administered surveys, which can provide a degree of privacy and confidentiality than in person.
With regard to the in-person interviews, the reason I do phenomenological interviews, often lasting seven or eight hours in length, is that through the length of conversation, some of those walls come down.