My argument is that I don't think that's what people hear when you ask them about "legal loopholes". And so the survey accidentally, implicitly ends up compari

My argument is that I don't think that's what people hear when you ask them about "legal loopholes". And so the survey accidentally, implicitly ends up comparing "legal tax avoidance" with "benefits fraud" which is apples to oranges. If the questions instead had been "working with your accountant to minimise you tax liability in legal ways" vs. "working with your benefits advisor to make sure you receive the benefits you legally qualify for", I suspect the result would have been different. If you want to write the article the author actually did, the more relevant question would probably have been "tax fraud" vs "benefits fraud". I'd be curious to know what those numbers are.