Beware of Media Selection Bias, Sensationalism

From an article by the Statistical Assessment Service (STATS) at George Mason University titled "The Media's Gas Problem":

"What people often don’t realize is that the media framing of scientific studies incorporates the journalist’s own perspective, whether the journalist realizes it or not. A dramatic example is the recent appearance of two scientific studies on fracking that provide a natural experiment on media sensationalism. 

Study One was critical of natural gas development; Study Two was supportive. How much coverage did each get in the mainstream media? The score: 

Study One by Cornell University researchers: 24 big-city newspaper articles and an NPR appearance.

Study Two by Carnegie-Mellon researchers: Two newspaper articles, one of them in a story primarily about Study One. 

The immediate takeaway from this story of dueling studies is that readers should be alert to the possibility that the media is emitting its own gas into this debate. The broader point is that the media’s treatment of scientific studies should be treated as a kind of rolling health scare, a structural imbalance based on a selection bias that is unlikely to change anytime soon."