Very interesting research, Frederic, very thought-provoking. I do see the quality part of the equation the most challenging in developing adequate metrics. The problem I‘m sensing here’, in Internet terms, is the existence of ‘digital tribes’, each of whom will assess quality based on their own values. For example, dedicated consumers of Fox News or Breitbart will assess quality in very different terms than those of the NY Times or Huffington Post. And even within these tribes you will have to work out distinctions. As you note, the NY Times spends, relatively, a lot more effort producing content than Huff Post, which includes a lot of regurgitation from other sites.
By ‘quality’ I’m not talking about the values expressed in the content — the differences here are obvious — but rather the actual structure of the content. The Fox News consumer wants bombast or strong emotional content, for example, while the more cerebral NY Times reader would find these values off-putting and a reflection of poor quality. And then there’s the obvious issues of balance, accuracy, and fairness in reporting, something that a NY Times reader would place a high value on, but is considered relatively unimportant to the Fox News reader.