Many tools that evaluate news or sources offer a single score: 73 out of 100, four stars, a green, amber or red light. It is easy to understand and easy to share. Honest News deliberately does not do this. Here is why.
A score hides the reasoning
A number is the end of an argument with the argument removed. If an article scores 62, which of its claims were wrong? Were they central or minor? Was the evidence missing, or contradictory? The reader cannot tell. A score asks to be trusted instead of showing why it should be.
A score averages things that should not be averaged
An article can contain three accurate background facts and one false central claim. Averaged, it looks "mostly reliable". In reality, the one false claim may be the only reason the article exists. Honest News evaluates each claim separately, so a true detail cannot hide a false headline.
A score cannot say "I don't know"
What score should a claim receive when the evidence is not yet available? 50 looks like "half true". 0 looks like "false". Neither is honest. A verdict of Insufficient Evidence says exactly what is the case: the question is open.
A score invites false precision
The difference between 71 and 74 suggests a precision no fact-checking process has. Scores also tempt people to rank sources and outlets, turning evidence questions into reputation contests.
What Honest News shows instead
For each claim, the Claim to Evidence engine shows:
- the exact quote from the text being checked;
- one of four verdicts: Verified, Refuted, Misleading or Insufficient Evidence;
- a confidence level, stated in words;
- a plain-language explanation of why;
- the sources, each with a note on whether its link was checked;
- any distortion notice, such as missing context or exaggeration.
It takes a few seconds longer to read than a number. That is the point. Evidence is meant to be looked at, not summarised away.