You can exercise judgment in choosing a topic: fair enough.

Within that topic, you can present all the facts that you know, or just the ones that support a certain narrative.

> Within that topic, you can present all the facts that you know, or just the ones that support a certain narrative.

Let's say you are reporting on the elevated rate of gun violence in Chicago compared to other US cities.

Which of the following facts do you include in your story? (For sake of argument, assume these are factually accurate statements.)

1. Chicago votes predominately Democrat.

2. Chicago has a higher poverty rate than other cities.

3. Chicago has a large black and Hispanic population.

4. Chicago has more gangs than other cities.

I think this is the wrong question to ask. I'd prefer to see reporters refrain from "connecting the dots". it's not a good fit for the daily/weekly news cycle.

there's a lot to discuss on this topic without "contextualizing". you can report on whether gun violence is up or down YoY, whether the percent delta is above the noise floor of the last few years, compare against trends in similar US cities, that sort of thing.

once you start including facts like 1, 2, 3, or 4, you are crafting a narrative, whether it is explicit or not. it's irresponsible to include this stuff unless you are prepared to make a detailed analysis of whether these factors actually predict crime, thorough comparisons against other cities where the same facts are true, possibly some regression analysis. most journalists lack the time and capacity to do this.

Why not all of those facts? It's only four short sentences, it won't take up much space on a printed newspaper let alone an online article. Throw in some more facts about American gun crime rates vs other countries for good measure, some facts comparing Chicago to other American cities, etc.

If it's your intent to report facts about gun violence in Chicago, and only that, then dropping those other facts detracts from what you're doing. It makes it look like you're insinuating possible causes. Like that voting Democrat leads to gun violence, or that poverty causes gun violence, or that hispanics and blacks perpetrate gun violence.

By mentioning certain remotely related facts, it looks as if you're hinting at giving an interpretation of the central facts.

If you want to convey facts about a topic like gun violence in Chicago, then stick to facts closely relevant to it. Like different ways of slicing up the data. You might show a density map of where in Chicago most of it has happened over a certain period, or demographics about gun violence.

Now if you want to explore a link between gun violence and something else like poverty, then you have to get your ducks in a row about that. Such a subject deserves to be treated by a sophisticated researcher with an expert background in statistics. If that researcher postulates some kind of link, they have to look at all the reasons why that might be false, not only true.

The idea is to report all the facts that at least a substantial minority of the public believe to be relevant, regardless of which way they swing politically. All of those facts qualify under this standard, and also the ones I suggested as well. If you leave out the part about Chicago being a Democratic stronghold city, half the American public are going to be yelling that fact at the article and concluding that the writer is biased. And if you leave out the comparison of American gun violence to other countries, you'll have the same accusations of bias coming from other readers.

If you want to preempt those accusations of bias, then you need to cover your bases and present all the facts that the public may believe to be relevant. Even those facts which you personally believe to be irrelevant distractions from the narrative you prefer.

What you're doing there is a form of populism: adjusting the biases hinted at your article to the anticipated biases held by the public, as to somehow pander to everyone at the same time.

That's not the same as avoiding biases.

The idea isn't to pander to the majority, but to include facts that at least a substantial minority believe are relevant even if most people think those facts aren't relevant. This would not make the majority feel pandered to, it would probably make most people believe the article isn't ideologically aligned with them which is the entire point. Democrats would say "why the hell is the article mentioning that Chicago is controlled by Democratic politicians? That's not relevant!" And Republicans would say "why the hell is the article mentioning that America's gun violence rate far exceeds that of other developed countries, that isn't relevant!" However in both cases they would have to concede that the article did include the facts they personally believe relevant (those facts which the other objects to) even though it also included irrelevant facts. This isn't pandering, this is impartially reporting the facts that some significant fraction of people might believe to be relevant while at the same time not conforming to their preferred narrative by omitting facts inconvenient to their narrative.

You can't pander to both at the same time. What you can do, and typically is done, is pander to one or the other by conforming to one narrative or the other, choosing or omitting facts according to the needs of one narrative or the other. That is pandering.

The facts you know are also selected subjectively, both by yourself and by the people who make you available to those facts.

Yes, and then bias is also a statistical phenomenon too, yadda yadda. But we all understand we're talking about the subset of bias that any specific actor can introduce due to ignorance or ulterior motives, and could in principle choose otherwise. Objectively unbiased reporting may be impossible, but intellectual honesty is possible, and should be required of people more.

Agreed. Even if one cannot achieve an ideal, there is no reason why one should not try their hardest to live up to it, instead of abandoning it as "all or nothing".

I don't think it's really that hard to report in a way that is unbiased enough to be useful. Just ask yourself, am I leaving out facts that my political opposite would think are relevant? But even just focusing on facts, however selected, would be better than what we have now which is a news media that decides what they want you to think first, and then reports or buries a story based on this.

It can be hard to even realize you're even being biased. As a professional journalist, just the fact that you're reporting on a region at all is a thumb your employer has put on the scale. It wouldn't be, were journalism evenly distributed across all conflicts, but there is no market for that kind of journalism, and so it isn't done.

That doesn't mean people should disregard journalism done from "journalism hot spots". But it's an important thing to keep in mind, and I think it bears especially on the question of how much emotional valence we should want invested in stories.

A view even allowing for a "political opposite" is already really limited for this purpose. What's the opposite of a christian anarchist? Is a soldier fighting a genocide its opposite, or is a pacifist? It's not that the model is useless but it's not very useful for this.

OurWorldInData.com is probably what unbiased "reporting" looks like, and obviously even they have to choose what to research and present.

At a deeper level, that site presupposes that quantitative data is a good way to interpret the world. (I don’t disagree, mostly!) Of course some things are easier to turn into numbers than others.

For some data there are multiples discordant sources, or critics directed to the unique source. Selecting (and therefore neglecting some) isn't neutral either. A real (non-disputable) consensus is very rare.