I wonder if your research takes into account people who just see headlines that confirm their bias and then share vs. people who actually click on the articles shared. I ask this because I’ve found generally the actual articles that qualify as “fake news” (a word that has virtually lost all meaning now) fall apart quickly upon examination.
The danger to me seems to lie in people seeing these articles with provocative headlines that confirm their bias, shared by a friend, acquaintance, or family member, and then mindlessly sharing themselves, thinking to themselves that its obviously correct because it makes sense to their worldview.
Also, it’s hard to square you saying that fake news isn’t a problem, when the definition of fake news is hard to quantify. Is it content that is flat out untrue? Or is it content that’s obviously misleading? Is it content that manipulates data to prove a point or articles that selectively leave out context?
I think if you just count content that’s flat out false, then I’d agree and say that the influence is probably small. People can be good at rooting out something that’s purely false. The far more insidious problem is the content that’s intentionally misleading, distorted, convoluted, inflammatory, painting a bad-faith picture of its target.
We’ve all seen the Facebook pages, shares, and shoddy websites that seem to garner hundreds of thousands of views, like, and shares. Where’s the data on that?