Here's the dilemma in assessing media coverage of the Trump presidency. Of course it is negative, more negative than coverage of previous presidents. But how much of that is warranted, and how much reflects a bias, or even hysteria, on the part of some in the news media?

Some clues are in a new report about Trump media coverage from the Pew Research Center. The study suggests that the bias of some media coverage is perhaps as much about the audience than the publication.

The report begins with an effort to compare coverage of the first days of the Trump administration with those of President Trump's three predecessors – Bill Clinton, George W. Bush, and Barack Obama. Of course, a lot of websites and other media outlets that are influential today did not even exist in Clinton's first term, or even Bush's. So the researchers chose to study a small group of media outlets – The Washington Post, New York Times, Newsweek, the ABC, CBS, and NBC evening newscasts, and the PBS NewsHour – that have existed the whole time, in order their coverage then and now.

They discovered huge differences over time. Assessing the traditional outlets' coverage of the first 60 days of each president, they found that Clinton's coverage was 28 percent negative and 27 percent positive (the rest being neutral); Bush's coverage was 28 percent negative and 22 percent positive; and Obama's coverage was 20 percent negative and 42 percent positive. Trump's coverage, by the same organizations? Sixty-two percent negative and five percent positive.

Of course, there were a lot of developments that merited negative coverage in Trump's first 60 days. But 62 percent to five? It's probably fair to say that is simply biased on its face.

The much larger part of the Pew study covers the first 100 days of the Trump administration, January 21 through April 30. The Pew researchers studied 24 news outlets, which they divided into three groups. The standard for dividing the groups was not the outlets' content but rather the "political makeup of their audiences," as measured by other Pew surveys. The groups were defined as 1) those whose audience "consists of two-thirds more who are right of center politically than left' 2) "two-thirds who are more left of center than right," and 3) "outlets with a more evenly distributed audience base."

The organizations with the most dominant left-leaning audiences were, in order, Vox, Slate, the New York Times, Politico, NPR, Huffington Post, Los Angeles Times, Washington Post, Buzzfeed, PBS, Business Insider, and MSNBC. The organizations at the top had the highest percentage of liberals and Democrats in their audience, but all skewed well left in audience composition.

The organizations with the more evenly distributed audiences were USA Today, CNN, NBC News, the Wall Street Journal, International Business Times, CBS News, and ABC News. All except ABC had an audience that leaned more left than right, but far less so than an organization like the New York Times.

The organizations with the most dominant right-leaning audience were Independent Journal Review, Fox News, Breitbart, Sean Hannity radio, and Rush Limbaugh radio.

The Pew researchers found that for media organizations with left-leaning audiences, like the New York Times, Washington Post, and others, 56 percent of stories had a negative assessment of Trump and the Trump administration, versus five percent that had a positive assessment, and 39 percent that appeared neutral.

For news organizations with a more balanced audience, 47 percent of stories had a negative assessment, while six percent had a positive assessment, and 47 percent appeared neutral.

Finally, for news organizations that had a right-leaning audience, 14 percent of stories had a negative assessment, while 31 percent had a positive assessment, and 55 appeared neutral.

One way to look at that is to note that Fox, IJR, Hannity radio, and Breitbart were more likely to publish a negative story about Trump than the New York Times, Vox, Politico, Washington Post, et al, were to publish a positive one.

In a broader sense, is anyone surprised that Sean Hannity's radio show is more positive toward Trump than the New York Times? Of course not. Look at Hannity's radio audience – 73 percent leaning conservative or Republican versus three percent leaning liberal or Democratic. The value of the Pew study is showing that the audience of a mainstream publication like the Times skews heavily in one direction – 51 percent who identify as liberals or Democrats versus 12 percent who identify as conservatives or Republicans – and therefore its coverage leans that way, too. News organizations are conscious of their audiences just like other businesses are conscious of their customers. Over the long run, the audience tends to get what it wants, presented in the way it wants it.

Pew discovered that one recipe for bias, in both directions, is focusing news stories not on policy but on the single person of Donald Trump. All the outlets, no matter their orientation, did that, Pew discovered.

"When reporting on any event, a reporter can choose any number of ways to orient the storyline," the Pew report said. "This study classified stories into one of two main frames: the president's leadership and character or his core ideology and policy agenda. Overall, journalists structured their narratives far more around President Trump's leadership and character than his policy agenda."

Nearly three-quarters of all stories, regardless of positive or negative, were focused on Trump's leadership and character – a better word might be "style" – while only one-quarter focused on his policies.

That, in effect, meant that a huge number of stories boiled down to whether Trump is bad or Trump is good. Not surprisingly, the outlets with liberal audiences said he is bad more often than the (fewer) outlets with conservative audiences.

There are lots of other details in the Pew study, about the number and type of sources the various groups rely on, about their story selection, and the like.

But the bottom line is that left-leaning readers, listeners, and viewers get what they want from their outlets of choice. So do right-leaning readers, listeners, and viewers. The New York Times audience expects something from the publication, just like Hannity's audience expects something from his program. And both of them get it.