How to Evaluate Information Effectively

Before evaluating a source, ask what you need it to do.

Consider:

• what your assessment brief and marking criteria require
• the question or claim you are investigating
• the kinds of evidence appropriate to your subject
• whether you need recent information, foundational research or both
• whether the source will provide evidence, context, data, theory, professional guidance or an example
• how important the information will be to your argument or decision

Many assessments require academic sources, but this does not mean that every useful source will be a journal article or book. Depending on the task, you might need legislation, statistics, datasets, professional guidance, policy documents, creative works, news coverage or other primary material.

A source should be judged against its purpose and the requirements of the task.

See Understanding source types for help recognising different forms of evidence.

Establish what you are looking at before deciding how much confidence to place in it.

Ask:

• Who created or published it?
• What type of source is it?
• Is it the original publication or a summary, report, repost or commentary?
• Has it been peer reviewed, professionally edited or formally approved?
• Is it a preprint, working paper or other version that may later change?
• Is the author writing from research, professional expertise, direct experience or personal opinion?
• Can you confirm the author, title, date and publication details elsewhere?

Different source types go through different forms of review. Peer review provides a level of academic scrutiny, but it does not guarantee that a study is correct, well designed or suitable for your work. Material that has not been peer reviewed is not automatically unusable, but its status should affect how you evaluate and describe it.

See:

• Academic and non-academic sources
• Primary, secondary and other types of source
• Peer review

Authority is not established by a job title, institutional name or professional-looking website alone.

Consider:

• whether the author or organisation has relevant knowledge or responsibility
• whether their expertise relates to the specific subject being discussed
• where the information was published and what review or editorial process was used
• why the source was created
• who its intended audience is
• whether commercial, political, professional or personal interests may have influenced it
• whether funding, affiliations or conflicts of interest are disclosed
• whether important context about the source is missing

Perspective or potential bias does not automatically make a source unusable. It may be part of what you need to analyse. The important questions are whether you understand the source’s position, whether its claims are supported and whether it is appropriate for your purpose.

When evaluating online information, investigate the author or organisation beyond the page itself. A domain ending such as .org, .gov or .edu, or a polished design, does not prove that the content is accurate or appropriate.

Lateral reading can help you check what independent sources say about an unfamiliar author, publication or organisation.

Do not judge a source only from its title, abstract, introduction or conclusion. Examine how its claims have been produced.

Depending on the source, ask:

• What evidence is provided?
• Where did the data or information come from?
• Is the method explained clearly enough to assess?
• Is the sample, case or dataset appropriate for the question?
• Are important groups, variables or perspectives missing?
• Does the analysis follow from the evidence?
• Are alternative explanations considered?
• Are uncertainty and limitations acknowledged?
• Are references provided, and do they lead to appropriate original sources?
• Has the source been corrected, retracted, challenged or superseded?

The relevant checks will vary between subjects. Evaluating a laboratory study, legal judgment, historical document, business report, artwork or professional guideline requires different disciplinary knowledge.

For important or specialised evidence, use the evaluation criteria taught within your programme and seek guidance from your tutor or supervisor where needed.

A source can be genuine and credible without supporting the claim attached to it.

When using a source as evidence:

1. Identify the precise claim you want it to support.
2. Find the relevant passage, data or finding in the original source.
3. Read enough of the surrounding material to understand the context.
4. Check the population, location, date, method and limitations where relevant.
5. Make sure your wording does not make the finding broader or more certain than the source allows.
6. Distinguish what the source found from your own interpretation.

Be particularly careful when a claim comes from a summary, search-result snippet, news report, social media post or AI-generated answer. Follow the information back to the original source wherever possible.

A list of references does not prove that an argument is well supported. You need to check how the sources have been selected, represented and connected to the claims being made.

Newer is not always better, and older is not automatically outdated.

Ask:

• Is the publication current enough for this topic?
• Has knowledge, policy, law, technology or professional practice changed since it was published?
• Is an older source important because it introduced a theory, method or debate?
• Does the source address your actual question?
• Does it concern the right population, location, setting or period?
• Is it written at an appropriate level for your work?
• Are you using the most suitable version or edition?

Currency matters differently across subjects. Recent evidence may be essential for current law, health guidance or technology, while older research may remain central to a theoretical, historical or creative topic.

Relevance also matters independently of quality. A well-conducted study may still be unsuitable if it examines a substantially different context from the one you are discussing.

Avoid relying on one source to represent an entire topic or debate.

Compare appropriate sources to identify:

• findings that are supported across several studies
• genuine disagreements or competing interpretations
• differences in methods, samples or contexts
• evidence that qualifies or challenges an earlier claim
• groups, perspectives or types of evidence that are under-represented
• uncertainty that has been hidden by a simple summary

Agreement between several sources can strengthen a conclusion, but the number of sources is not enough by itself. Several publications may repeat the same unsupported claim or draw on the same underlying evidence.

Citation counts show that a publication has attracted attention. They do not establish that it is reliable, correct or relevant. Counts also vary by subject, publication date and database.

See Recognising and addressing bias for more help considering perspectives, assumptions and omissions.

Quick checks can help you decide whether a source deserves closer attention. You might:

• scan the title and abstract for relevance
• check the author, publisher and publication date
• identify the source type and publication status
• look for references, data and an explanation of the method
• check whether a correction or retraction has been issued
• investigate an unfamiliar website or organisation through other sources

These checks help you filter results, but they do not replace reading and evaluating important sources in full.

Evaluation frameworks such as CRAAP, SIFT, PROMPT or lateral reading can provide useful prompts. They are not scoring systems that automatically establish quality, and you do not have to use a named framework unless your programme requires one.

See Evaluation frameworks explained for examples and guidance on choosing an appropriate approach.

For help investigating misleading online claims, see Spotting misinformation and disinformation.

Evaluating an original source and evaluating an AI-generated answer are related but different tasks.

A reliable source does not automatically make an AI answer reliable. The tool may:

• invent or misidentify a source
• attach a genuine source to the wrong claim
• misrepresent what the source says
• remove important context or limitations
• select evidence supporting only one position
• produce an incomplete or unbalanced synthesis

First evaluate the original sources. Then check how accurately and fairly the AI output has used them.

See Evaluating AI outputs for the complete process.

After evaluating a source, decide whether to:

• use it as evidence
• use it for background or context only
• use it with an explanation of its limitations
• compare it with stronger or conflicting evidence
• find a more appropriate source
• exclude it from your work

Evaluation is not about finding perfect sources. It is about making informed decisions, recognising limitations and being able to explain why the evidence is suitable for your purpose.