Skip to Main Content

Evaluating Online Information

How to determine if information found on the internet is credible.

Strategies for Different Types of Sources

The same basic strategy is used for evaluating all sources: fact check and read laterally.  The following are some more specific tips for the different types of sources you will encounter.

News sites

Newsguard is tracking AI-generated news' websites

Use the resources below to do your own fact-checking.

How to Spot Fake News (YouTube video from FactCheck.org)

 

The following infographic can also help!

Fake news infographic. Readable text PDF version attached.

Source

Website

Nearly every company, organization, institution, media outlet, retailer, and many individuals have a website.  Evaluating websites is where lateral reading and the SIFT methodology are invaluable to help you decide if you should trust the information you are seeing and reading.

Ask yourself the following when you encounter a website:

  • How did you find the page?  How you located a site can give you a start on assessing the validity as a resource.
  • What is the site's domain?  The origin of the site can provide an indication of the site's mission or purpose.  Here are some of the most common domains and their meanings:
    • .org :An advocacy web site, such as a not-for-profit organization.
    • .com : A business or commercial site.
    • .net:A site from a network organization or an Internet service provider.;
    • .edu :A site affiliated with a higher education institution.
    • .gov: A federal government site.
    • .wi.us :A state government site, this may also include public schools and community colleges.
    • .uk (United Kingdom) : A site originating in another country (as indicated by the 2 letter code).
    • ~:The tilde usually indicates a personal page.

Blog Post or Social Media

  • A person's blog or social media posts can be more difficult to evaluate but it is still essential to do so!  You need to find out who is behind the information that you see posted there.   Again, lateral reading can help you here.
  • Social media actually adds more complexity to assessing information because it can actually cause information to find you even when you didn't ask for it.  The algorithms that work behind the scenes on social media sites use your likes, shares, and clicks to funnel information that it thinks you'd be interested in directly to you without you even entering one search term.  This article from the journal In the Library With The Lead Pipe discusses this concept.

ChatGPT/Other Generative AI tools

  • ChatGPT and other generative AI text, image, and video content generators present information in a "sanitized" manner which makes it very difficult to evaulate the source of the output or content produced.  So, what can you do?  Fact-check and use lateral reading.
  • Here's a way to approach it:
    • Break down the information.  Take a look at the output from your prompt and find a specific, searchable claim.
    • Then do your lateral reading!  Open a new tab on your web browser and look for supporting information.
    • Next, think critically about what assumptions are being made both in your prompt and in the generative AI output.  Who knows things about this topic and where can you check to find out?
    • Lastly, you need to make a judgment call.  What is true and what is inaccurate or misleading? You may need to dive deeper in some of your sources.

This short video from the University of Maryland libraries shows you how!

Data

3 Ways to Spot a Bad Statistic (TED talk)

Assess the data using the following strategy:

  • Investigate the Source
    • Was data collected by individual researchers or an organization?
    • Does the person(s) who reported and collected the data have an affiliation with an entity that may bias the work? 
    • Can you identify the original source of the data?
    • Why was the data collected in the first place? 
  • Investigate the Impact 
    • What is the impact of the data producer?  How often is their work used or cited?
    • What is the impact of the data itself?  How often is it cited? 
    • How is the data used in the research?  How does their use compare with yours?
    • Do peers have objections to the measure or data producer?

What about scholarly sources?

  • Just like with internet information, you must know your source when it comes to scholarly literature.  Start your quest using the Oxy library website since it directs you to reputable databases and other online resources.
  • The reputation of the journal or author of an article can also help you assess validity of a source.  Search for the author on the internet or use Oxy library's subscription to Scopus to check the impact factor of a journal or to verify an author, their affiliations, and what they've published.  Web of Science covers all disciplines, not just science!
  • If a journal article has been peer-reviewed (the article has been read by other experts in the field before publication), then that adds another layer of credibility to an article.  Keep in mind that they are many other types of information that are not peer-reviewed but are still trustworthy such as technical reports, U.S. government websites (e.g. National Library of Medicine, National Institutes of Health), and the Smithsonian. 
    • Note: As of this writing (2/3/25) major changes are being made to US government websites.  Please double check the information that you get from these sites using other sources or look for archives of pages prior to January 28, 2025.
  • Read critically!  Are the claims made in the article concordant with the results?  Check out the references section of the paper to see who the author is citing and then read the papers to see if they are accurately representing the article they are citing.
  • Read this article for a critical analysis of a research study in action!