Mental model

Evaluating Evidence

Systematically assess information quality to form sound judgments and make better decisions.

Discover

A new health supplement claims to boost energy levels based on testimonials from three people. Which question would reveal the weakest evidence?

Testing your evidence radar

Build a sharper filter for everyday claims.

Understand

Understand

Evaluating evidence means checking if information actually supports a claim before accepting it as true. You look at where the evidence came from, how much was gathered, and whether it was collected fairly. For example, a friend saying a restaurant is great based on one visit is weaker evidence than reading reviews from hundreds of customers over several years. Ask this:

Full explanation

Full explanation

First, assess the source. Evidence from peer-reviewed research, established institutions, or experts with relevant credentials carries more weight than anecdotes, social media posts, or sources with conflicts of interest. A medical claim backed by a randomized controlled trial published in a reputable journal is more reliable than one supported only by testimonials.

Second, examine the quality and quantity of evidence. Single examples or personal stories ("my aunt tried this diet and it worked") are the weakest form. Studies with large, diverse samples, repeated across different research groups, provide much stronger support. When reading that a product works for most users, look for how many people were actually tested and whether the sample reflected the population you're in.

Third, watch for selection bias and cherry-picking. People naturally seek information that confirms what they already believe. A political news site might only share stories favorable to one side, or a company might highlight only their best reviews while ignoring negative ones. Strong evidence considers all relevant data, not just what supports a preferred conclusion.

Finally, consider how directly the evidence connects to the claim. Studies on rats don't automatically apply to humans. Research from one context might not transfer to another. Good evaluation matches the evidence to the specific claim being made, in the specific context where it matters.

Research

Research

Nickerson (1998): Confirmation bias operates as a "ubiquitous phenomenon" causing people to seek, interpret, and remember evidence in ways that confirm pre-existing beliefs, leading to systematic errors in evidence evaluation. [1]

Stanford Encyclopedia of Philosophy: Critical thinking involves "consideration of any belief or supposed form of knowledge in the light of the grounds that support it, and the further conclusions to which it tends," requiring explicit standards for evidence quality. [2]

Kahneman (2011): People rely on fast, intuitive thinking (System 1) that accepts familiar and coherent stories as true, while slower, analytical thinking (System 2) needed for proper evidence evaluation requires deliberate effort and attention. [3]

Limitations

Limitations

Evaluating evidence perfectly is cognitively demanding and often impractical for daily decisions. Time constraints, limited expertise, and information overload mean we must rely on heuristics and trusted sources. Complete neutrality may be impossible since everyone has existing beliefs that shape how they interpret evidence. Additionally, some domains (art, ethics, personal relationships) rely on forms of knowing beyond empirical evidence. The process can also be exploited by those who learn to present weak evidence in ways that appear methodical, using technical language or selective data to create an illusion of rigor.

Try it

Synthesize

Choose a pattern from the guide, then pick an action to try with it.

Which pattern stands out?

What will you try?

Choose a pattern above to select an action.

Sources

Sources

Try it

Check your understanding

A viral post claims that leaving onions in your room prevents flu because "my whole family does it and we haven't been sick in years." What makes this the weakest type of evidence?

Show the guide's explanation

Answer: It's a personal anecdote without comparison

Anecdotes are the weakest evidence because they represent a single case without controls. The family might have other protective factors, fewer exposures, or simple luck. Strong evidence requires systematic comparison between groups with and without the intervention to isolate the effect.

Which piece of evidence would strengthen a claim that a new study technique improves test scores?

Show the guide's explanation

Answer: A randomized trial with 200+ students showing score gains

Randomized controlled trials with large samples provide the strongest evidence because they rule out alternative explanations and demonstrate consistent effects across many people. Endorsements and testimonials, even from experts, reflect personal opinions rather than systematic evidence.

You're evaluating whether to believe a headline that states "New Study Shows Coffee Shortens Lifespan." What should be your FIRST step in assessing the evidence?

Show the guide's explanation

Answer: Find the original research and examine its methods

Before accepting any headline about research, trace it to the source. Was it a large, long-term study? Was it peer-reviewed? Did it control for other factors? Headlines often oversimplify or distort research findings. Understanding the actual evidence (sample size, study design, effect size) matters far more than popularity or personal habits.

Keep exploring

Find another idea for the decision in front of you.

The complete Reframo library is free to read. Explore another guide whenever you are ready.

Evaluating Evidence | Reframo