Why should I trust the Factual Grade?

The Factual automatically calculates how informative an individual article is. This is based on four factors:

• The diversity and quality of its sources
• The neutrality of tone in the article
• The expertise of the journalist on the topic based on their writing history
• The site reputation based on historical grades of every article scored for the site

Because the calculation is automated and independent of human involvement, criteria are consistently applied across articles and sources. More on the rating algorithm.

Also, because grades are specific to an article and not a publication, grades vary within a publication. For example, here is a medium grade article and here is a high grade article, both at the New York Times.

How do you use the Factual Grade?

The Factual automatically calculates how informative an individual article is.

• A grade above 74% indicates a well-written article that is highly likely to be informative. Such articles provide extensive sources, maintain a relatively neutral tone, and are authored by a reporter who has apparent expertise in this topic. That said, there may still be bias.
• A grade below 50% means the article is less likely to be informative. These articles may still be worth reading but should be cross-checked with higher graded articles.
• The average grade for articles is 56%, with 95% of articles scoring between 28% and 85%.

How reliable is the Factual Grade?

The Factual grades 10,000+ articles a day and gets better each day as it evaluates more articles, sources, and authors. Some known issues:

• We do not always factor in multimedia links, such as tweets, videos, and photos, into a article's analysis.
• Authors who write for different sites are treated as different authors.
• We are unable to identify authors in some news sites.
• Non-English sites are not yet graded by The Factual.
• Sites behind a subscription wall like The Wall Street Journal and Bloomberg may not be accessible to grade.

Why is my favorite site or journalist graded below average?

Many popular journalists write articles with very few links or are highly opinionated. This may be because of their writing style or policies of their publisher/editor. Such articles will likely receive a below average grade from The Factual. While this may seem unfair, it is incumbent on journalists (and publishers) to link to good sources and moderate their opinions, so readers can reach their own conclusions on topics they care about.

As the American Press Institute says journalism is a "discipline of verification." Online, it's easy to quote and link to sources so journalists who do so often will rightfully score more highly. The Factual assesses this approach--and more--so that you, the reader, can reach your own conclusions on the news.

A forthcoming update will credit journalists who are cited most, which typically correlates with those who break stories.

Where do political bias ratings come from?

Political bias ratings for each source come from a combination of sources:

• The non-partisan site AllSides
• The non-partisan site Media Bias Fact Check

There is no uniformly accepted way to judge bias of a news site, as bias is relative to some baseline and people cannot agree on what that baseline is. Bias information is included to help you understand how a story may be framed, or even why a story is being reported on. And this is why bias ratings do not factor into The Factual Grade and are provided merely as context for how a story may be framed.

Why are all comments anonymous?

To minimize echo chambers, all user comments are anonymous so people can speak honestly without professional backlash or without feeling pressure to say what others expect of them. Also, you cannot see how many upvotes a comment has received, except your own comments, to reduce voting based on popularity.

How does The Factual moderate content?

To avoid flame wars, The Factual seeks to highlight the highest quality comments first rather than the most popular comments. We determine comment quality based on several signals such as length, tone of writing, and, notably, how likely the commenter is to be knowledgeable based on "Respects" they've earned. Comments with profanity are removed.

What are Respects?

Respects are The Factual’s points system that helps identify readers who are likely to be knowledgeable. While everyone should have a voice, those that are knowledgeable are likely to have better insights about issues on average. Hence we seek to identify and promote comments from such users.

How do you earn Respects?

You can earn Respects in two ways:

1. For every day you read at least one article you get a Respect. This pattern of reading consistently suggests a reader is knowledgeable. Note that reading multiple articles on a single day still only gets you one Respect for that day.
2. Upvotes from readers who already have 5 or more Respects. This means you are getting upvotes from people who are likely knowledgeable and hence suggests your comment has merit.

When are Respects counted?

As you receive upvotes your Respects total will update within a few minutes. For reading an article each day your Respects will update nightly. Note that Respects next to a user’s anonymous handle on a specific comment remain unchanged as an indicator of the Respects the user had at the time they made that comment. Also, only Respects gained in the last 90 days affects comment quality scores so that long-standing users are not always favored compared to newer users.

Why can’t I see how many upvotes (Respects) a comment has?

Showing the Respect count encourages popularity-based voting. We want people to vote independent of other people’s actions. However, you can see how many Respects your own comment received.

Why are upvotes called Respects?

We wanted a verb that indicates admiration for a comment even if you disagree with it. It’s a way to agree or disagree with grace.

How can I trust you?

We do not produce any news ourselves and are not affiliated with any political group or cause.

The company has a geographically diverse team to reduce our own inherent biases.

We are transparent about our process for selecting stories and our algorithm's scoring model. We respond to valid criticism and make changes if necessary.