That's literally the point he's making! The terms "flawed" and "zombie" reflect positive identification of dodgy data, so of course when the data isn't availabl

That's literally the point he's making! The terms "flawed" and "zombie" reflect positive identification of dodgy data, so of course when the data isn't available, they are less likely to apply, hence why: > This finding alarmed him, too: it suggested that, without access to the IPD — which journal editors usually don’t request and reviewers don’t see — even an experienced sleuth cannot spot hidden flaws.