Chapter 4: Evaluating information
Most of the time, you won’t be able to determine the reliability of a claim or statement based on how it is covered in a single source. To get a better understanding of your topic and ensure that the information you’re using is reliable, try to find better coverage. This can take many forms: you might find better coverage through sources that are more detailed, varied, authoritative, or recent. Even if you think your original source is good, verifying what you’ve found with additional sources will strengthen your research.
Rather than assuming the information presented within a single article provides a full picture of your topic, read multiple sources and examine how they support or contradict your original source. This is called lateral reading, going beyond your original source to explore how the same topic is discussed by others. The primary purpose of reading laterally is to quickly determine whether the original information you found is true. It’s more than just fact-checking, though; this process can also uncover better information if the source you started with is not as reliable as you had hoped.
Here is an example of reading laterally in action. You may find a news article that asserts “coffee reduces your risk of getting cancer.” Seek out additional sources to determine whether there is any research that backs up that claim. Open up a new browser window and search for the phrase “coffee and cancer risk.” This may reveal if other news sites are reporting the same information, and how they have interpreted the results. Sometimes, multiple news sources will post or broadcast the same story word-for-word. Just because a story is shared widely doesn’t mean that it is accurate, and it doesn’t tell you where the data came from. Keep searching to find a better source. “After verifying, refuting, or otherwise contextualizing the data in question, you’d return to the original text and continue reading.” If you find that the information is being misrepresented or if you cannot find anything to back up the claim, don’t use your original source. Instead, seek out new sources on your topic that do a better job of supporting their claims.
Navigating around paywalls
In the past, we encouraged using Wikipedia or Google for quick fact-checking and investigation. However, lateral reading may require more extensive searching because many reliable sources can only be accessed through subscriptions. As an ISU student, you can use the library’s article indexes and databases to access many of these subscription-based materials for free.
Let’s say you want to verify a news story; you find another article on the topic but the newspaper’s website won’t let you read the entire article without subscribing. In cases like this, check if the ISU Library subscribes to that newspaper through a database like Access World News, or if there is access directly, as for the New York Times. The best way to find this information for many local and major newspapers is to check the Newspapers Library Guide.
Alternatively, your professor may provide a link to an article that you need to read, but that leads to a paywall. You can use the search techniques for tracking back a citation from Chapter 3 to find the article you need, or use a tool like LibKey Nomad. LibKey Nomad is a browser extension that enables one-click access to the full text of articles from publishers, Google, Wikipedia, and more, without needing to start at the library website. Visit the LibKey Nomad Library Guide to learn more about this tool.
This process works for any source you want to learn more about. Another benefit of reading laterally is that you may encounter additional sources that will better inform your project and support your argument.
Interference from algorithms and generative Artificial Intelligence (AI)
Another reason library search tools may be useful for finding better coverage is to avoid the heavy bias built into the search algorithms used by Google and other web search engines. These tools change your search results based on your personal search history, online activity, and location and are affected by algorithmic bias. Unlike the other types of bias we mentioned in the last section, algorithmic bias can be difficult to detect: algorithms may appear to be objective because they are based on data, but the ways that data is gathered, used, or manipulated often reflects the biases of the algorithm’s creators and users. “In search engines, for example, algorithmic bias can create search results that reflect racist, sexist, or other social biases, despite the presumed neutrality of the data.”
Your results are skewed by what Google thinks you want, based on what it “knows” about you. This may change what results are displayed but, more importantly, it determines the order they’re listed in. If you and another person search for the same thing, what is displayed first for each of you may be different. Don’t let Google decide for you what is and isn’t relevant for your needs. Use a tool that is less likely to use your browsing history to inform what it retrieves for you, such as DuckDuckGo. Everyday tips like clearing your search history and cookies can also lessen these algorithms’ effect on your search results.
You know now that interference from algorithmic bias and the artificial intelligence (AI) behind these search engine algorithms affects your research in unseen ways. With recent advancements in generative AI such as ChatGPT, Bard, and Dall-E, we are beginning to notice these technologies impacting research at all levels. In addition to being affected by algorithmic bias, generative AI has the ability to create convincing yet false statements. As the use of generative AI becomes more widespread, it is getting harder to find and distinguish between reliable sources with accurate information and AI-generated content full of misinformation.
For example, when Google first released their generative AI, Bard, it claimed that the James Webb Space Telescope (JWST) took the first pictures of an exoplanet. This was quickly proved false by space enthusiasts who pointed out the first picture of an exoplanet was actually taken in 2004—years before the JWST was launched. This is one of the reasons why you absolutely have to evaluate your resources critically before you decide to use them in your research.)
Whether you are searching for sources on Google or using AI just to help you generate ideas, you still need to check for inaccurate information and be cautious of algorithmic bias.
- News Literacy Project. (n.d.). Expand your view with lateral reading. Retrieved August 16, 2021 from https://newslit.org/tips-tools/expand-your-view-with-lateral-reading/ ↵
- Heick, T. (n.d.). What's the difference between lateral reading and vertical reading? TeachThought. Retrieved from https://www.teachthought.com/literacy/how-res-ding-different-future-literacy/ ↵
- Butler, W.D., Sargent, A., and Smith, K. (2020). Algorithmic bias. In Introduction to College Research. Retrieved from https://introtocollegeresearch.pressbooks.com/chapter/algorithmic-bias/ ↵