Searching and Finding

Authenticating Information

The strength and weakness of the internet as a research source is just how much information there is: a badly-phrased search can drown you in irrelevant, misleading or unreliable results. This is why some research has shown that companion reading can actually backfire if people don’t have effective search skills.[1]

Where we look online depends a lot on what we’re looking for:

  • If you want to hear people’s personal experiences and opinions, social networks and forum sites are full of them.
  • For a practical topic like learning how to fix a bike or looking for a good place to eat, a video site like YouTube or TikTok can be a good choice.
  • Wikis are useful to explore and learn more about something. You can start with Wikipedia, of course, but there are also lots of wikis on specific topics.

For other topics, though, you’ll want to turn to a search engine like Google or DuckDuckGo. Canadian youth report making more sophisticated use of search engines than in the past: two thirds of participants’ in Phase IV of MediaSmarts’ Young Canadians in a Wireless World survey said they used advanced search engine tools, compared to just over a third in Phase III, and a similar number scanned the full first page of search engine results before clicking on one compared to half in Phase III.[2]

Keep in mind that search engines aren’t actually sources of information – they’re gateways to sources that have to be read and evaluated. You should always go to the actual source and judge it for yourself.

Both search engines and social networks are designed to deliver what’s most relevant to you – which they determine by tracking what you read, what you watch, what you click on, what you purchase and where you go online (among many other things), then feeding that information to an algorithm that ‘decides’ what content you get to see.

The best way to avoid this (in addition to varying your sources of information) is to limit the ways in which search engines and social networks collect data about you, so that they’re less able to build a profile. To do this, you can:

  • Make a habit of using Incognito or Private Browsing mode in your browser.
  • If you have a Google account, go to My Google Activity and make sure Web & App Activity, Timeline and YouTube History are turned off.
  • Change your settings to limit tracking in browsers such as Chrome, Safari, Firefox and Edge.
  • If you use an iOS device, limit tracking by going to Settings, then Privacy & Security, then Tracking, and turning off the toggle Allow Apps to Request to Track.
  • Use a non-tracking search engine such as DuckDuckGo.
  • Install a blocking plugin, such as PrivacyBadger, or a blocking app, like DuckDuckGo or Blokada.
  • Set the ad tracking and feed preferences in your social network and search engine accounts to not deliver customized content.

Learn how to fine-tune a search

Search engines work best when you include as many specific terms as possible. For information on the movie Grand Hotel for example, you would want to include the word “movie” (to avoid hits referring to actual hotels called the Grand Hotel); if you wanted to get information on the 2016 movie rather than the more famous 1932 release, you might add the term “2016.”

You get the best results by including terms that are as specific as possible to what you’re searching for. If you want information on U.S. President Ronald Reagan’s “Star Wars” space defence plan, for instance, you would add the term “Reagan” to avoid results that only refer to the movie Star Wars.

While using more specific terms is generally helpful, though, it’s also important to make sure that your search terms won’t give you results tilted towards a particular point of view. If your search is in the form of a question, it can wind up giving you the answer you want. For instance, searching for “Is walking better than running?” gives you results that are either neutral or in favour of walking, while searching for “Is running better than walking?” gives you the reverse.[3] Similarly, including specific terms connected to the answer can give you misleading results. If you’re trying to find out whether NFL football ratings went up or down as a result of players kneeling during the U.S. national anthem, searching for “NFL ratings down” will give you results that say they went up and searching for “NFL ratings up” will give you results that say they went down.[4]

To make a search as effective as possible, use the “Advanced Search” option available in most search engines. (In Google you can access this feature in “Settings”.) This will give you additional search options such as filtering by language, by date or by limiting a search to one website in particular.

To save time, you can also put some of these directly in your search term:

  • Put all search terms in quotes. That tells the search engine to only show you results that include those specific words. If you search for cats dogs without quotes, you might get results that include cats but not dogs, or dogs but not cats. If you search for “cats” “dogs” you will only get results that include the word “dogs” and the word “cats.”
  • You can leave out results that include a specific term by using the minus sign (so adding -dogs will leave out any results that include the word “dogs”).
  • You can search only one website by adding “site:” and then the Web address (so adding site:mediasmarts.ca will give you only results from that site).
    • In the case of some very well-known sites, such as Wikipedia, you can get similar results just by adding the name of the website to your search.
  • You can leave out results from a website by adding the minus sign and then the Web address (so adding -mediasmarts.ca will leave out results from that website). This is very useful when you want to find out about a website, because it will only show you what other sources say about it.
  • You can search only sites from one country by adding “location:” and then the name of the country (so adding location:Canada” will only give you results from websites based in Canada).
  • You can search for specific types of files by adding the word “filetype:” and then the file extension (so adding filetype:pdf will only give you Adobe PDF files).
  • You can search for a word in the title of a website using the word “intitle.” For instance, searching for intitle:hippo will give you only results with the word hippo in the page title.

You can also combine two or more of these tools. For example, if you search for house hippo” –mediasmarts.ca you’ll see websites that include the words “house hippo” but not MediaSmarts’ website.

Remember that you can redo and improve your searches. If you find a lot of irrelevant results, for instance, you might find some words or phrases that you can exclude.

If you have a Google account, you can also make your own custom search engine. This lets you search just the sites you’ve chosen, so you can make a library of sources that you know are reliable. (Research has found that people who curate their information diet, through tools like bookmarks or a custom search engine, are less likely to be fooled by misinformation.)[5] You can even make several different search engines on different topics. To find out how, see the MediaSmart tip sheet Make Your Own Custom Search Engine.

Evaluating search results

Making a good search query is just the first step. We also need to judge which of the results is most likely to be accurate and relevant. This is why click restraint – waiting until you’ve looked over the results before clicking or tapping one – is an essential information sorting skill.[6]

To do that, it helps to understand how search engines sort results. Advertisers can pay to have their page appear higher in the results; look for “Ad” or “sponsored.”[7] There is a whole industry called search engine optimization that is about “gaming” search algorithms. There’s not necessarily anything wrong with this – most legitimate businesses and organizations do it to make sure their content is “findable” – but unfortunately, it’s the dishonest ones who are usually the best at it, and there’s no way for you to tell from your search results which ones have been manipulated in this way.[8]

The most effective form of search engine optimization is taking advantage of – or making - data voids. That’s when searching for a topic or phrase leads gets just a small number of results. For instance, if you wanted to spread a rumour that pickles were poisonous, you might make videos and websites titled “Pickles are poison,” then tell people to search for that phrase.[9]

  • That’s why disinformation spreaders will often encourage you to search for a particular phrase, knowing that it will lead you to their message. Even if you do want to learn more, use a neutral search like “pickles health effects” instead.

For “organic” or unpaid results, Google ranks pages on the criteria of beneficial purpose and page quality. Beneficial purpose prioritizes sites and pages that are likely to be helpful to users, as opposed to clickbait and other sites that are purely commercial and provide little value to users; examples include sites that share information or opinions, that sell products and services, that entertain and so on.[10]

Page quality is determined according to four factors: relevant expertise, experience (including personal experience, in topics such as hobbies), authoritativeness and trustworthiness.[11] Sites are considered low quality if they have misleading titles, no listed authors or authors with no relevant expertise or a negative reputation, distracting content like intrusive ads, and harmful or deceptive content. The highest page quality standards are applied to what’s called “your money or your life” content such as medical or legal information, news articles or other topics that can have a large impact on users’ lives.[12]

While this method makes sense, and is certainly better than not taking things like expertise and authority into consideration, it can also be “gamed” by people using search engine optimization (SEO) techniques.[13] It can also lead to unintended consequences: for example, sites are also ranked more highly if they include what’s called supplementary content like relevant links, sidebar tips, background information and so on.[14] As a result, online recipe sites almost always now have an anecdote or explanation before getting to the recipe itself, making them less useful.[15]

Once you have your results, skim the first dozen or so to see which ones are the most relevant to you and come from reliable sources. On Google, you can click or tap the three dots next to the result. That will give you general information about the site – usually pulled from Wikipedia.

  • If you’re using a search engine that doesn’t have that feature, you can open a new tab and check Wikipedia yourself.

It may seem counterintuitive, but it’s actually a good idea not to spend too much time on your search.[16] Instead, stop once you’ve found a few good sources, open them in new tabs, and only go back to the search if none of those turn out to be helpful.

  • Search results on Google may also include a Knowledge Panel, which is a summary – often taken from Wikipedia – of what the search engine thinks you’re looking for. Don’t assume that the Knowledge Panel or the search results give you the answer by themselves. Make sure to follow the links that you’ve determined are most likely to be useful.[17]

Go beyond the Web (into the Deep Web)

Search engines don’t index the entire Web — a lot of the information that’s available sits in databases or behind pages that require users to log in.

For example, the catalog of books at your local library is most likely not indexed by Web search engines and you can only search it by using their own search engine. The same goes for many other collections of content such as the Getty Images collection of stock photos, or the collection of public domain music in the Internet Archive.

Before going straight to a search engine, therefore, ask yourself if there’s a more specific source that you should try first.

Here are some examples of Deep Web sources you may find useful (note that some of these may contain advertising, collect user data or charge for some services):

Don’t rely on autosuggestions

Many search engines will also give you some suggestions for search strings as you start typing your keywords. For example, if you start typing “Jacques”, the search engine may display “Jacques Cartier” as one of its suggestions. This can be very useful in cases where you’re not sure about the spelling of something but, because these suggestions are based on what other people have searched for or your previous searches, they can also be misleading.

Don’t use voice search, mobile search or chatbots for important topics

More and more of us are using mobile devices such as tablets and smartphones for searching the internet,[18] using voice-operated search tools,[19] or both. These can be very convenient, but there are a number of reasons why you shouldn’t use them for anything more important than finding restaurants or a weather forecast.[20] Not only is it difficult to use any of the search tools described above, but these searches have some biases built in that you might not want: for example, mobile search results generally prioritize results that are near you or are happening right now.[21]

Chatbots and “AI search engines” should not be used for important topics: research has found that as many as half their answers are inaccurate in some ways, and more than a third provide incomplete or harmful information.[22] If you do use one, either choose one that lists the sources it used to get your answer or ask it to provide sources. Be sure to check that the references really exist[23] and make sure they’ve been summarized effectively.

[1] Blair, R. A., Gottlieb, J., Nyhan, B., Paler, L., Argote, P., & Stainfield, C. J. (2023). Interventions to counter misinformation: Lessons from the Global North and applications to the Global South. Current Opinion in Psychology, 101732.

[2] MediaSmarts. (2023). “Young Canadians in a Wireless World, Phase IV: Digital Media Literacy and Digital Citizenship.” MediaSmarts. Ottawa.

[3] Caulfield, M. (2017). "Avoiding Confirmation Bias in Searches" in Web Literacy for Student Fact-Checkers. 
<https://webliteracy.pressbooks.com>

[4] Tripodi, F. (2018) How Trump Voters Decide Who to Trust.  Augmenting Humanity.

[5] Bateman, J., & Jackson D. (2024) Countering Disinformation Effectively: An Evidence-Based Policy Guide. Carnegie Endowment for International Peace.

[6] McGrew, S., & Glass, A. C. (2021). Click restraint: Teaching students to analyze search results. In Proceedings of the 14th International Conference on Computer-Supported Collaborative Learning-CSCL 2021. International Society of the Learning Sciences.

[7] Ham, C. D. (2019). Why Is This First? Understanding and Analyzing Internet Search Results. Journal of Educational Research and Practice, 9(1), 400-412.

[8] Bradshaw, S. (2019). Disinformation optimised: Gaming search engine algorithms to amplify junk news. Internet policy review, 8(4), 1-24.

[9] Bradshaw, S. (2019). Disinformation optimised: Gaming search engine algorithms to amplify junk news. Internet policy review, 8(4), 1-24.

[10] (2022) EAT, YMYL, & Beneficial Purpose: What Do Google’s Quality Standards Mean for Search? Digital Strategy One. < https://www.digitalstrategyone.com/eat-ymyl-beneficial-purpose-what-do-googles-quality-standards-mean-for-search/>

[11] V., K. (2023) Google’s Search Quality Evaluator Guidelines: What Does the E-E-A-T Update Mean? Express Writers. < https://expresswriters.com/google-search-guidelines-released/>

[12] (2022) EAT, YMYL, & Beneficial Purpose: What Do Google’s Quality Standards Mean for Search? Digital Strategy One. < https://www.digitalstrategyone.com/eat-ymyl-beneficial-purpose-what-do-googles-quality-standards-mean-for-search/>

[13] Warzel, C. (2022) The Open Secret of Google Search. The Atlantic.

[14] V., K. (2023) Google’s Search Quality Evaluator Guidelines: What Does the E-E-A-T Update Mean? Express Writers. <https://expresswriters.com/google-search-guidelines-released/>

[15] Broderick, R. (2023) The end of the Googleverse. The Verge.

[16] Caulfield, M. (2021) Information Literacy for Mortals. Project Information Literacy.

[17] Beirich, H. (2020) Google’s Knowledge Graph Is Rife with Misinformation and an Easy Tool for Online Radicalization. Global Project Against Hate and Extremism.

[18] (2014) Mobile Search Advertising Around the Globe: 2014 Annual Report. Marin Software.

[19] “The Voice UI Has Gone Mainstream.” Tech.pinions, https://techpinions.com/the-voice-ui-has-gone-mainstream/46148

[20] Andersson, C. (2022). Smartphones and online search: shifting frames in the everyday life of young people. Information and learning sciences, 123(7/8), 351-370.

[21] (2017) Mobile Search: It's Different. Forrester.

[22] Angwin, J., Nelson A. & Palta R. (2024) Seeking Reliable Election Information? Don’t Trust AI. Proof News.

[23] Chiu, J. (2023) ChatGPT is generating fake news stories attributed to real journalists. I set out to separate fact from fiction. The Toronto Star.