The Nature of Health in the United States and the World, Hist 3180/ENVS 3980: Research strategies

List of resource types for use in planning the types of materials you want to look for.

1. Think about the disciplinary perspective(s) from which you want to find materials on your topic and make sure to search databases and library catalogs that are strong in those disciplines. For instance, a paper on the Exxon Valdez oil spill in Alaska in 1989 can be approached from the perspectives of geography, the environment, history, politics, law, economics, and so on. In addition, a general overview covering all of these perspectives could be provided. Each approach can be researched using different databases.
2. Make a list of types of materials you would like to find or search for and be sure to use databases, library catalogs, print finding aids, and/or archival finding aids that include or specialize in those materials.
   • Books
   • Articles in scholarly encyclopedias are a great place to start a research project. They provide an overview of a topic, cross-references, and a bibliography of the most important scholarly materials on that topic.
   • Review articles (to be distinguished from book reviews) can be very useful resources for serious research projects. They are written for specialists in the field. They review the literature on a topic and synthesize the results; thus, the bibliography is more extensive than that of an encyclopedia article.
   • Scholarly journal articles
   • Popular magazine articles
   • Newspapers
   • Dissertations or Master's Theses
   • Government documents (federal, state, local, international)
   • Government records (federal, state, local, international)
   • Technical documents
   • Manuscripts
   • Maps
   • Media (DVDs, CD-ROMs, Video, etc.)
   • Websites
   • Conference papers
   • Working papers
   • Materials published in languages other than English or those translated into English
   • Statistics
   • Primary sources
   • etc.
3. Take advantage of database trials currently available.
4. Use Compass, MaineCat, or WorldCat to identify and to request materials that Bowdoin College Library does not hold.
5. Visit a library nearby to use its print collection and/or electronic databases.

Selected search strategies for use in library catalogs and databases.

1. Keyword Searching. Conduct a keyword search in a database using either heuristically developed search terms or a list of carefully developed search terms, depending on the database and your needs. If permitted in the database, you may want to combine terms using Boolean AND, OR, or NOT. Find materials of interest and note the relevant subject headings, descriptors, topics, etc. for further use in Keyword Searching or in Subject Searching.
2. Subject Searching. Subject search in a database. A good way to develop a list of terms to be used in Subject Searching is by Keyword Searching. If permitted in the database, you may want to combine terms using Boolean AND, OR, or NOT.
3. Author Searching. Identify an author who has published in the area of interest. Conduct an author search to find other materials by that author. Use Google or search at the author's home institution to find the author's CV, which might contain other materials or materials not yet published.
4. Footnote Chasing (a.k.a. Backward Citation Searching). Find a key book, article, dissertation, or other item of interest, then follow the notes and/or bibliographic entries. In so doing, one follows the research network backward in time. This can be done based on a print or on a full-text electronic version of the item. Some of the databases in the box "Citation Searching in Subscription Databases" below provide bibliographies that one can use to footnote chase.
5. Citation Searching (a.k.a. Forward Citation Searching). Find a key book, article, dissertation, or other item of interest, then find articles or books that cite it. This enables one to follow the research network forward in time. To be an effective search technique, the key source should have been published long enough ago that it has had time to be cited. See boxes below.
6. Browsing. Find a book of interest that is shelved using the Library of Congress Classification (LCC) scheme. Go to the shelf where the book is located (or would be, were it on the shelf). Browse other materials near the book of interest. Compass can also be browsed: find a book of interest in Compass; use "Virtual Browse" OR "Browse CBBcat Shelf". This strategy allows for serendipitous finds.
7. Journal Run. Identify an important journal in the area of interest. Then search or browse through all volumes in the relevant years. As Bates notes, "this approach exploits Bradford's Law: the core journals in a subject area are going to have very high rates of relevant materials in that area".

* From a given article, click "Times Cited in this Database" to view citing articles. Citation searching is available through "Cited References" (blue bar at the top). See Cited References, EBSCO.

Author Searching, Browsing (what she calls "Area Scanning"), Citation Searching, Footnote chasing, Journal Run, and Subject Searching are from Marcia J. Bates' classic article, "The Design of Browsing and Berrypicking Techniques for the Online Search Interface." Online Review 13 (October 1989): 407-424.

How do you know when you have done a fairly comprehensive search? (Note: There is no such thing as a 100% comprehensive search.) Here are some things to do and to look for.

1. You have narrowed your research question appropriately.
2. You have revised and refined the vocabulary that you use to search.
3. You have used a variety of relevant search strategies. (See "Checklist of Search Strategies for Catalogs and Databases" box.)
4. You have searched a variety of relevant databases, perhaps including
   • disciplinary and multidisciplinary databases
   • databases for related disciplines
   • databases for relevant publication types. (See "Checklist of Resource Types" box.)
5. Your results lists contain highly relevant publications. To be confident that your search is not too narrow, i.e. that you are retrieving a high percentage of the highly relevant publications, the results lists should also include some (but not an overwhelming percentage of) less relevant ones. (For definitions and a technical discussion, see Precision and recall in Wikipedia.)
6. You regularly see the same highly relevant publications appear over and over again in your results lists.