This Harvard Study of negative coverage gets cited a lot as “proof” of the MSM’s bias against Trump. But this argument is a bad interpretation of the study, based on two, major logical-flaws. I wrote this a comment response to someone, but I think it deserves its own post. I think my reasoning is correct. Please let me know where I am wrong. The study being referenced is talked about in this article.
You keep referencing the Harvard Study. I didn’t want to go into this into too much detail because I’m tired of explaining this over and over again. Your reliance on this study is based on two, major logical flaws. I will explain how.
Your reference Harvard study shows that you do not understand what bias, or what it does or does not imply. Bias by itself has no bearing on credibility, and cannot be used as a singular feature to assess it. Case in point:
- Outlet A that largely reports accurate, positive news about X, has high standards of journalism, and retracts articles when shown to be false
- Outlet B that largely reports accurate, negative news about X, has high standards of journalism, and retracts articles when shown to be false
- Outlet C that largely reports positive fake news about X, has little to no standards of journalism, wilfully engages in misinformation and does not retract anything.
- Outlet D that largely reports negative fake news about X, has little to no standards of journalism, wilfully engages in misinformation and does not retract anything.
- Outlet E that largely reports both negative and positive fake news about X, has little to no standards of journalism, wilfully engages in misinformation and does not retract anything.
Outlets A and B are biased but credible. Outlets C and D are biased and not credible. Outlet E is non-biased and not credible.
Therefore bias alone does not imply credibility; it may be bias but it can also be something else, which leads us to your second, major logical-flaw: you assume that reality must always reflect 50% positive and 50% negative coverage. This is a ridiculous assumption. A true distribution exists and it may not be a fair (in the mathematical sense) one. Case in point:
- Person A engages in horrible conduct (any sort of horrible conduct you can think of).
- Person B is just a regular human being who has made some mistakes, but is largely a good person.
- Person C largely engages in negative conduct, but has also done some good things (e.g., a mafia don who gives free food to his neighborhood).
Now take each of those persons above, and substitute them for X in the example outlets are shown above. What do you see? The definitions of A, B, and C shows the true distribution of their behavior. The kinds of reporting from A, B, C, D, E can influence a person’s perception of their behavior. What does this mean? That you cannot solely consider those outlets in isolation, using bias, as a metric for credibility. You must consider outside, credible, corroborating sources.
Now in the case of Donald Trump, there is an extremely large amount of corroborating evidence, both from his actual, public, verifiable behavior, and from the comments of his close associates, which point to a general consensus that he is an odious man. He does multiple things on a daily basis that objectively display his lack of fitness for the Office of President; especially his inexperience, incompetence, ignorance, intemperance, lack of intelligence, lack of poise, and complete lack of principles and responsibility.
Comparing the reporting of MSM outlets with corroborating evidence of Trump’s words, behavior, and conduct leads us to the inevitable, logical conclusion that MSM coverage is a close reflection of the true distribution of Trump’s behavior, and one cannot use the mere fact that there is an imbalance in the sentiment of coverage to allege that there is a bias.