For sure, interviews of the famous produce unreliable information–that is, information that may, or may not, be correct–and one reason is because “[i]nterviewing the famous. . . is a game, a ritualised conflict with rules that are understood: `The participants. . . know that this is a transaction in which we both hope to get something more than we intend to give. . . . No body pretends to befriend anyone so there can be no question of betrayal. It’s perhaps a somewhat hard-headed transaction but not I am sure a morally ambiguous one'”. [Jenny Turner, quoting Lynn Barber, in the London Review of Books, 6/5/2014, p. 21.]
Gore Vidal, famous and extensively interviewed, agrees: “You have to become an explainer. You have to make up your mind before [the interview] . . . that you’re going to make [certain] points]. . . . I will then have thought it out in my head how I’ll lead the conversation. It doesn’t make any difference what they ask; you just go right on. `Yes, that’s interesting’, and go right on to the point you were going to make”. [John Mitzel, Steven Abbott, and the Gay Study Group, “Gore Vidal: The Fag Rag Interview, in Richard Peabody and Lucinda Ebersole, eds., Conversations with Gore Vidal (University Press of Mississippi, 2005), p. 22.]
Donald Trump lies to reporters–and admits it. “In The Art of the Deal, he says that whenever a reporter is out to get him, he tries `to frame a positive answer, even if that means shifting the ground’. He admits that some people might call this `lying’, but he prefers the phrase `truthful hyperbole’.” [Deborah Friedell, “Tycooniest”, London Review of Books, 10/22/2015.]
The larger, and also valid point is that all types/modes of asking and answering (polls, focus groups, etc.) produce unreliable information. It’s not that every response to every interview/question is incorrect; some answers in some circumstances probably are correct, but when you only have answers it’s impossible to know which, if any, are correct or incorrect. The only way to know is to check or verify answers with data from two or more non-asking sources; such as observations and documents. Interviewers and all other askers do not have data from non-asking sources; all they have are answers; all they have is unreliable information.
For a complete statement of the unreliability of answers to questions, see my book, The Problem with Survey Research.