Evaluating and Refining Your Search Strategy

As we have mentioned before, as you research, the steps of the research process will be repeated multiple times, ideally with improvements each time. To do that, you’ll need to evaluate your search results so you can think about how to make them better. You should not plan to just take the first article that pops up in the first search that you do. Your first search is extremely unlikely to be your best search!

Refining Your Search Strategy

Sometimes, a quick scan of your search results will be all it takes to realize that your search needs refining. You might need to adjust your search terms, your filters, or both!

If You Got Too Few Results, You Might Try:

• Making your keywords broader, such as searching for “social media” instead of “Instagram”
• Using “OR” to link together several synonyms or related terms and grouping with parentheses
  • Examples: (“grade point average” OR “academic achievement”), (Hispanic OR Latino), and (dogs OR canines)
• Removing some filters—only do this if the filters you remove weren’t important to your information need
• Making your date range filter wider, such as including the past 10 to 20 years instead of the past 5 years
• Breaking your topic into pieces, knowing that you may never find one perfect source. View the video below to learn how to do this.

If You Got Too Many Results, You Might Try:

• Making your keywords narrower, such as searching for “elementary school students” instead of “children”
• Using quotation marks around exact phrases, as described in the “Google and Other Search Engines” chapter
• Searching in a subject-specific database instead of Search@UW or Google
• Adding additional filters—only do this if the filters you select makes sense for your information need
• Refining your research question to be more precise (see the “Understanding Your Information Need” chapter for info on this)

If Your Results Just Aren’t What You Expected:

• If your search results don’t quite fit the topic you wanted, revisit the list of alternative keywords that you developed (in the “Keywords” chapter) and switch out your search terms for some of those. Sometimes the best keywords are terms that you hadn’t thought of yet! Try skimming the titles in the results list to see if you can spot any good terms to try searching with.
• If your search results are mainly books, but you were expecting articles, you might need to make your research question more specific—see the “Understanding Your Information Need” chapter.
• If the sources seem outdated, try making your date-range filter narrower. For example, include the past 3 to 5 years instead of the past 10 years; but don’t make it too narrow. It takes time to publish academic sources!
• If you can’t figure out why your search results aren’t what you expected, ask a librarian for help.

Selecting the Best Sources

After a couple rounds of using the above steps to improve your search, your search results should be looking pretty good! However, you still need to pick the most useful sources from the list. The algorithm that decides the order of the results won’t know exactly which sources are best for you and your information need. That’s why it’s not safe to just assume that the first few sources in the list will be the best ones to select.

Evaluating the Titles of Your Sources

Up until the now, you have just been skimming the titles of the sources in your results list to get a general idea of whether they are on the right track. Now it’s time to read the titles a little more carefully. Let’s say that your research topic is about the Russian Revolution. Let’s even say you specifically searched for information about the role of music in the Russian Revolution. Your search results included the articles:

• “‘Like the beating of my heart’: A discourse analysis of Muscovite musicians’ letters during the Russian Revolution”
• “Garbage cans and metal pipes: Bolshevik music and the politics of proletariat propaganda”

The titles tell you about the subject matter of the article and about how the author approaches the subject matter. You might not know exactly what discourse analysis is, but you can guess that you can do it to letters and that you should pay particular attention to it when the author mentions it in the article. Looking at the other title, you would know to watch out for very different words and concepts. It is up to you to decide which of those articles would be more valuable to your research, or if it would be valuable to read both! Before making the final decision, you’ll want to read the abstract, which we will cover in the next section.

Keep in mind that the convention within some academic disciplines to have a pretty long title separated by a colon usually follows a predictable pattern. The text to the left of the colon serves as a teaser, or as something to grab a reader’s attention (remember that the author is likely not trying to grab your attention, so you may not find these teasers particularly effective—though it is probably packed with phrases that would entice someone who already studies the topic). The information to the right of the colon typically is a more straightforward explanation of what the article is about.

Evaluating the Abstracts of Your Sources

Once you have identified a number of source titles that sound good, it will be worth it to evaluate each source a little further to determine how useful it will be for your information need. This way, you will save yourself the time of reading an entire source that turns out to not be what you need.

Not all of your sources will come with abstracts, but when they do, pay close attention. An abstract is a summary, usually one paragraph at the beginning of an article that serves to encapsulate the main points of the article. It’s generally a pretty specialized summary that seeks to answer specific questions. These include:

• The main problem or question
• The approach (how did the author(s) do the work they write about in the article?)
• The shiny new thing that this article does (to be published in an academic journal you often need to argue that you are doing something that has not been done before)
• Why people who are already invested in this field should care (in other words, you should be able to figure out why another academic should find the article important)

The abstract often appears in database searches and helps scholars decide if they want to seek out the full article.

That’s a whole lot to accomplish in one paragraph.

As a result, authors often use specialized jargon to convey complex ideas in few words, make assumptions of prior knowledge, and don’t worry much about general readability. Thus, abstracts are generally dense, and it’s not uncommon to read through an abstract and not have a clue about what you just read. This is a good place to reread, highlight, underline, and look up words that you don’t know. You still may not have a firm grasp on everything in the abstract, but you should treat the key terms in the abstract like parts of a map when you see them in the main text, leading you to treasure: understanding the main argument. We will come back to this in the “Reading Scholarly Sources” chapter.

Final Thoughts

Is it worth it to spend time refining your search and evaluating your results? It’s tempting to just declare your early search results to be “good enough.” Understanding the complex vocabulary used in article titles and abstracts can be difficult and frustrating. We think it is worth it to pick the best sources rather than the “good enough” sources, because it will make your research project or paper much stronger. It will take some effort, but you can do it!

Attributions

This chapter contains material adapted from: 

• “One Perfect Source?” by North Carolina State University Libraries, used under a CC BY-NC-SA 3.0 US License
• Reading Games: Strategies for Reading Scholarly Sources by Karen Rosenberg, in Writing Spaces: Readings on Writing, Volume 2, used under a CC-BY-NC-SA 3.0 license