The problem with this study is that the data wasn't controlled for being about a president vs. a candidate. There's going to be a number of mundane stories men

The problem with this study is that the data wasn't controlled for being about a president vs. a candidate. There's going to be a number of mundane stories mentioning the president (i.e. bills being signed, travel, meetings, etc.) that are outside the scope of what's also being reported on a candidate. The result is that any organization that is reporting on the president will inevitably mention the president in articles that are less biased, skewing the results comparison.