Skip to main content
note
This page was translated from the German original, partly by machine. Some passages may read awkwardly or contain inaccuracies. When in doubt, please read the original.

Evaluating sources: the criteria

When the quick SIFT check is not enough, for example with a source you need for an important decision, an essay or an article, a closer look is worthwhile. Six criteria help with this.

The six core questions​

1. Authorship: Who says this?​

Is there a named author or a responsible organisation? Is there a legal notice / masthead? A reputable source does not hide. Anonymity is no proof of untruth, but it is a reason for caution.

2. Expertise: Does the person actually know about this?​

Does the author have a demonstrable qualification or experience in this field? Beware the transfer fallacy: a famous physicist is not thereby an expert on nutrition. (See also the appeal to authority in the chapter on fallacies.)

3. Reputation: What is the source's reputation?​

Is the source cited and taken seriously by other reputable bodies? Does it have an editorial process, corrections, a culture of admitting mistakes? Reputation is best checked laterally, that is via other sources, not via its self-presentation.

4. Currency: Is the information still up to date?​

When was the content published or last updated? On fast-moving topics (technology, medicine, statistics) information ages quickly. On timeless topics, by contrast, an old date is no flaw.

5. Independence: Who profits?​

Who funds the source? Does it pursue an economic, political or ideological self-interest? Advertising, sponsoring and conflicts of interest need not automatically deceive, but they must be transparent. A lack of transparency is a warning sign.

6. Evidence: What supports this?​

Are claims backed up with verifiable sources? Do the references lead to reputable primary sources, or in a circle back to the author themselves? A strong claim without evidence is just an opinion with a loud voice.

Remember

A claim is not the same as evidence. The burden of proof lies with the one who asserts something, not with the one who doubts.

The benefit question: Cui bono?​

A single question bundles much of this: Who benefits from this information? (Cui bono = to whose advantage?)
Who wants me to believe this, and what do they gain if I do?

It is the same question with which the detective in a crime story gets on the trail of the culprit: who benefits from the deed? Whoever has a motive moves into the circle of suspects. With a piece of information it is no different. This does not make anyone automatically guilty: even those who profit can be right. But a recognisable self-interest is a reason to look more closely.

This question exposes advertising dressed up as news, interest-driven „studies“ and political agitation. It is no substitute for checking, but it is a good start.

Two well-known mnemonics​

CRAAP test​

From the world of libraries comes the CRAAP test (developed in 2004 by Sarah Blakeslee, CSU Chico). The acronym sums up five of the criteria above:

  • Currency: how up to date it is
  • Relevance: relevance to your own topic
  • Authority: authorship/authority
  • Accuracy: accuracy/evidence
  • Purpose: purpose/intent

Source: Meriam Library, CSU Chico (CRAAP test).

CRAAP is thorough, but it has a weakness: you work through the source „from the inside“ and check precisely the features that are easy to fake. So: first SIFT (check from the outside), then if needed CRAAP (evaluate in detail).

PICK: choosing the source for the task​

Once credibility is settled, PICK helps with the question of whether the source is the right one for your purpose:

  • Purpose: what kind of source is this, and does the genre fit my question?
  • Information: is the content relevant and useful to me?
  • Creation date: is it recent enough?
  • Knowledge-building: does it really advance my understanding?

Source: SBCC Library, SIFT & PICK (libguides.sbcc.edu).

In short

SIFT settles: Can I trust the source? PICK settles: Is it the right one for my task?

Mini checklist to take away​

  • Author/origin known? Legal notice present?
  • Expertise in this field demonstrable?
  • Reputation of the source checked (laterally)?
  • Date recent enough for the topic?
  • Funding/interests transparent?
  • Evidence present and traceable to the primary source?
  • Cui bono: who benefits from this?