This search strategy graph provides a visual illustration of the information retrieval process:

The basic steps in the search process are outlined below:

1. Define your topic.

   Before you can begin your search, you must clearly define a researchable topic. This means that you must have specific questions or problems that need to be addressed.

   Inappropriate or unsearchable topic:
   "I want to know about AIDS in Africa."

   As you can see, this topic is too broad. There's no way you can approach this topic in MEDLINE or other databases because you haven't defined a clear, researchable question. Ask yourself what it is that you really want to focus on. Be specific, and write the topic in narrative form.

   Appropriate, focused, and searchable topic:

   "I want to find information related to the incidence and prevalence of HIV and AIDS in teenage girls in rural and urban Ghana, or in other African nations."

   This topic is more appropriate because it specific enough to use to create a searchable question. Notice that this topic identifies a specific aspect of AIDS/HIV (incidence and prevalence), a specific population (teenage girls), and a specific region (Ghana). This topic also takes into consideration that the literature from Ghana may be slim, so you acknowledge that you may have to branch out to other nations.

   Now that you've focused your idea more clearly, you can move to the next step in the search process.

2. Identify key concepts outlined in your topic.

   Write out your topic and underline, hilite, or circle the main concepts. This process will help you determine exactly how you can approach the search strategy.

   "I want to find information related to the incidence and prevalence of HIV and AIDS in teenage girls in rural and urban Ghana, or in other African nations."

   Now rank the key concepts in order of importance:
   incidence and prevalence of HIV and AIDS
   Ghana
   other African nations
   females/girls
   teenage

   It's important to understand that your specific topic may or may not be well represented in the literature. You may have to come at the topic from a different point-of-view, or use related literature to address your question.

   Depending upon the amount of literature on your topic, you may have to modify your focus. For example, you may discover that there is a lot of literature on HIV/AIDS in Ghana, but not much specific to teenage girls. Perhaps you can modify your focus to include females of all ages. If this approach does not find more information, then you may have to modify your focus again. For example, you may discover that there is not much literature specific to Ghana, so you may want to include other African nations.

   Consider this example:
   If you are looking for information regarding the "self-concept of teenage boys who have had liver transplants," chances are that you'll find little or nothing in the literature. So, you have to change your point-of-view. Try searching for the self-concept of any age group. Or, try searching for the self-concept of any gender, any age group, that have undergone any kind of transplant.

3. Determine how the key search terms are to be combined together.

   Using the boolean AND and OR, think about how the components of your topic should be combined.

   Remember though, searching is a step-by-step process. FIRST search for the keywords or terms separately, THEN combine them together. Your completed search would look something like this:

   

   You'll notice above that the results of the search is 7 articles. Try broading the topic a bit to see if the retrieval increases.

   Below, we've added additional western african nations to the topic by using the Medical Subject Heading, "africa, western". We've also exploded [exp] "africa, western" to incorporate all the nations in that region of the world.

   See that we now have retrieved 173 articles:

   

4. Refine search with additional terms or limiting factors.

   Age groups (teenagers) and gender (girls) are check tags or limiting factors.

   This means that they are not included in your initial search statement as keywords or medical subject headings (MeSH). Instead, they are added on at the end using the limit set option. Other check tags or limiting factors include language (English, French, etc.), publication dates (1990-99, 2000, etc.), local holdings, journal subsets (cancer, dentistry, or nursing journals), or publication types (clinical trials, review articles, etc.)

   After we limit set to the "adolescence" age group and "female" gender, we have 61 remaining articles:

   

5. At this point, you should view and print relevant citations, or further revise the search strategy to find new citations.

Tips for understanding and interpreting search results:

• Look at subject headings from a relevant article, and try using it in your search.
• Look at references from relevant articles. What subject headings have they used? Can you use similar subject headings to improve your retrieval?
• Analyze why certain citation were retrieved VS. were not retrieved.
• Try a variety of approaches and combinations.
• Ask a professional searcher (librarian) for assistance.

• Why did my search find what it found?
• How do I know if I've found everything on this topic?
• What other approaches should I try?
• Am I using the right database, time period, etc.?
• Am I using the right textwords or subject headings?