Go to next section, Starting Points for your Research

Reference Sources

This section of the guide presents a framework for undertaking forced migration research.  It describes resources that can be used 1) as starting points for research, 2) to help expand research, and 3) to support research.  While specific titles are highlighted to help jumpstart the research process, tips for conducting database searches are also included to facilitate retrieval of additional resources beyond those listed here.  To ensure that readers proceed through this guide with the same basic knowledge about search strategies, I begin with an overview of factors to consider when preparing to search, and then I review several methods of searching.  These methods will be referenced regularly throughout Parts 2 and 3 of the guide.

1. Search Essentials: Setting the Stage for Research 

Research in the library or on the Internet is an iterative process that involves undertaking a series of searches to retrieve a set of information sources that can be read, analyzed, distilled and applied to a particular research question.  As human knowledge has expanded, so have the number of finding tools developed to facilitate access to it.  While every researcher has a different method for diving into this repository of human knowledge, there are certain facts that are important to know about the nature of forced migration information and the structure of finding tools that can ensure greater success.  

Information Formats

Fact: Human knowledge is recorded both informally and formally, in a variety of formats.  For research purposes, the most common types of information formats (also referred to as “secondary sources”) in the forced migration context include books, journal articles, and reports, in either print or electronic media.  Finding tools usually target specific information formats, i.e., library catalogues describe books, journal indexes provide access to scholarly articles, documentation databases house international document collections, etc.  Exceptions include digital libraries that house multiple formats because they focus on a certain subject discipline or issue.  The Internet is useful for locating reports as well as citations to books and articles; the full-texts of the latter can generally be accessed online but not for free.

Implications for research:   To locate different information formats, it will be necessary to search more than one finding tool accordingly.

Information Producers

Fact:  The introduction highlighted the many different actors within the forced migration community who work with, study, assist, protect, promote the rights of, and raise awareness of issues concerning forced migrants: academic researchers, the staff of international organizations, humanitarian aid workers, government decision-makers, policy personnel in both think tanks and governmental agencies, etc.  As a result, the range of information produced is diverse and extensive, and it is distributed through both formal and informal channels.  Information that is unpublished or semi-published – for example, policy documents, conference proceedings, reports, working papers, and the like – tends to be distributed informally, and is referred to as “grey literature.”  Generally, grey literature is more difficult to track down because it circulates less widely than commercially published materials, and as a result, it is less likely to be indexed or catalogued by mainstream information services.  Fortunately, the producers of grey literature are increasingly making this type of information available online. 

Implications for research:  Formal publications tend to cite other formal publications, while informal publications often reference other informal publications.  Comprehensive research requires regular searching and monitoring of both types of distribution channels.

Multidisciplinary Field

Fact:  Forced migration is of interest to a wide range of social scientists, legal scholars, and medical researchers.  In his review of articles published in the Journal of Refugee Studies, Black (2001) identified the following as the top ten disciplinary backgrounds of authors: political science, anthropology, sociology, psychology/mental health, socio-legal studies, history, international relations, health studies, geography, and education.  He goes on to speculate that “the majority of the scholarly literature on refugees remains in broadly disciplinary or policy studies journals, rather than in the more specialist journals…” (p. 61) such as the Journal of Refugee Studies. 

Implications for research:  This “high-scatter” effect is typical of multidisciplinary fields.  Without a central point of access, research becomes a question of magnitude – conducting a journal article search requires looking in five disciplinary indexes rather than just one – making it that much more challenging to keep up with the latest research developments.

Libraries versus the Internet

Fact: The Internet has not only made a wider range of information products more easily accessible such as news and reports, but it also has introduced entirely new formats such as blogs, web sites, and multimedia resources.  Yet despite the wealth of information on the Internet, it is false to assume that everything is online or that research can be limited to the online environment.  Hundreds of millions of books currently exist in print.  While information about them can be found online, the texts themselves often cannot.  A number of digitization projects are hoping to change this reality, most notably Google Books.[7] But for now—and no doubt years to come—books will continue to be retrieved through libraries.  

Implications for research:  Researchers should adopt a hybrid approach that encompasses both print and online resources.

Information Access

Fact:  The ultimate objective of research is to retrieve full-text information.  Each researcher will have different options available to him or her, in terms of Internet or library access:  Online access at the office?  Through a library or Internet café?  At home?  Access to a good library collection close by?  As noted above, both the Internet and libraries offer important resources for research.  Cost is invariably a factor for researchers, and may be a deciding point in one’s choice of a finding tool or information product.  There is a widely held perception that everything on the Internet is free.  While a multitude of free resources are available online, many key research tools such as journal databases require either a subscription to search their collections, payment for access to full-text content, or both.  The advantage of libraries is that they assume the subscription costs of many research databases and journals which in turn are made freely available to the user.  The only requirement is that the user be physically present in the library and that s/he cover print or photocopy costs.  While some libraries restrict access to a certain type of patron, many others do not.  For example, public libraries are open to all-comers.  University libraries often allow members of the public to use their collections, but may require a small user fee to borrow books.  National libraries are usually available to people with bona fide research needs.  Specialized collections such as the Refugee Studies Centre library collection are open to anyone. 

Implications for research:  If you are not affiliated with a university or institution with an in-house library, look for libraries in your area and inquire about their access policies.[8] If you do not have online access, you may be able to search the web at the library or an Internet café.[9]  If you are based in a developing country, your library or institution may be able to take advantage of special arrangements that have been negotiated with content publishers in order to access full-text resources on a complimentary or reduced rate basis.[10]  Another alternative is to take advantage of document delivery services.  For example, the Refugee Studies Centre library will—subject to copyright restrictions—supply photocopies of certain documents for a small fee that covers postage and paper costs.[11]  In addition, some fee-based services[12] will go one step further and conduct a literature search for you as well as provide copies of relevant texts.   

2.  Methods of Searching 

Today, many users associate searching with a simple box into which one or two keywords are entered.  While this type of “quick search” is ubiquitous on the web, other search options are on offer from finding tools such as library catalogues and journal databases (see Part 3 for a listing of additional finding tools).  As noted above, it is helpful to keep in mind the kind of information source housed in the database you are searching: a web page? a journal article? a report? a bibliographic record?  The nature and structure of an information source will give you a hint as to the elements that might be searchable.  For example, title keywords, author names, subject terms, words within abstracts, or words anywhere within a text all represent searchable elements of information sources.  For this reason, it is useful to distinguish between field searching (often presented as a search form) and full-text searching. 

This section provides a basic primer on these types of searching with the caveat that there is no standardization in this area.  Therefore, when confronted with a new database or information repository, it is always advisable to become oriented with its features by reading the “Help” section in order to enhance your chances for conducting a successful search. 

Boolean Searching

Because “Boolean operators” are used in both field and full-text searching, they serve as a useful starting point when considering methods of searching.  Boolean is a means of logically combining search terms using AND, OR, NOT to narrow or expand one’s search results.  Usually, two terms entered into one field or into a quick search box are assumed to be joined by AND:  >>conflict Sudan<< is the same search as >>conflict AND Sudan<<.  Combining terms with AND will limit search results because the requirements for a match are more stringent.  For example, searching for >>refugees women Iraqis<< requires a database to locate records with all three terms present to make a match, whereas entering the single term >>refugees<< only requires one term to be present.  Combining terms with OR will expand search results because a match can be made if either one or the other term is located: >>Iraqis OR Palestinians<<.  The OR operator is particularly useful for synonym searching and for locating word variations:  >>nationality OR citizenship<<, >>women OR woman<<.  The NOT operator allows you to omit a term from search results.  This is useful if you notice that certain search results appear that are not relevant because of the presence of a certain word:  >>trafficking NOT drug<<. 

Variations on Boolean: 

1) Some databases support wildcard searches.  A wildcard (*) search is a quick means of performing an OR search, and is used to locate word variations.  For example, rather than entering >>stateless OR statelessness<<, enter >>stateless*<< to achieve the same result.  >>Humanitarian*<< will search for “humanitarian,” “humanitarians,” or “humanitarianism.”  Many web search engines automatically perform a variant of wildcard searching called stem searching.  For example, enter >>refugee<< and the plural >>refugees<< will also be retrieved. 

2)  As noted above, two terms entered into a search box will usually be interpreted as an AND search.  If the two terms form phrases like internally displaced, involuntary resettlement, literature review, and Amnesty International, many databases will allow you to place quotation marks around the terms to require the system to locate them adjacent to one another and in the order entered:  >>“internally displaced”<<, >>“involuntary resettlement”<<, >>“literature review”<<, >>“Amnesty International”<<.

Boolean is a strategy for crafting precise search queries.  However, it is often deemed overly technical for the non-specialist.  Web search engines have popularized search in such a way that many either do not accept Boolean operators or they do not require them to be entered directly.  Users simply type in various terms, and search results are presented according to the underlying algorithms that define the search engine.  These usually incorporate Boolean in such a way that the user does not need to understand it to benefit from it.  While this may be sufficient for certain types of searches, serious research requires a basic understanding of how Boolean functions since it is routinely referenced in many of the databases that forced migration researchers will encounter during the course of their work.

Field Searching

Databases that store information sources often design search engines to allow users to search various elements within either a bibliographic record (as in a library catalogue) or a full-text document (as in an article database).  These elements are generally referred to as fields.  Typical fields that can be searched within both library catalogues and journal databases are title, author, subject term, and keyword (example).  These are described in more detail below.  Other examples of fields might include abstract, geographic terms, language, publisher, and publication date.  The search interface is usually presented as a form that allows users to enter terms in each specific field they wish to have searched.  Often, users can also indicate how they would like to combine the search terms they have entered by selecting from a list of Boolean operators (usually AND, OR, NOT) (example).  This method enables users to perform much more precise and focused searches than can be done with web search engines.  What follows is a description of the four most common fields, with tips on what to enter into each.

Title field:  Generally, when a title field search is performed, the system will look for matches to keywords within titles, in any order and ANDed together.  For example, >>refugee law international << will locate The Refugee in International Law by Guy Goodwin-Gill, along with other books whose titles contain those three terms.  However, some systems may interpret terms entered into a title field literally; in other words, they will only make a match when the title words begin with >>refugee law international<<, in the order presented and with the terms adjacent to one another.  This particular search would most likely yield zero hits as a result.  In these systems, entering >>refugee in international<< will produce a match.  (Notice that the initial “the” in the title can be omitted.)  Alternatively, if >>refugee law international<< produces zero matches in a system, try adding AND in between each of the terms.  Note also that some systems may offer separate fields for searching titles generally and journal titles specifically. 

Author field:  In most systems, entering the last name of an author into the author field will be sufficient to make a match.  However, with more common names, it may be necessary to enter a first name as well, in order to reduce the number of hits.  While most systems accept author keywords in any order and assume a default AND in between, some systems are more rigid.  In these cases, the proper form for entering names may be either >>last name first name<< (in the exact order entered) or >>first/last name AND first/last name<< (first and last names in any order but joined by an AND).  Hyphenated family names can prove problematic in some systems.  Therefore, if you enter >>harrell bond<< in an author field and receive zero matches, try entering the name with a hyphen in between (>>harrell-bond<<).  Note that author fields usually accept both personal and corporate authors. 

Subject field:  Also referred to as the descriptor, index term, or subject heading field.  It is distinguished from the keyword field, into which any term can be entered, because it is usually linked to a thesaurus.  A thesaurus is a controlled vocabulary list that presents designated terms for subjects that may be described in many different ways or for variant word spellings.  The following examples illustrate the utility of standardized terminology when searching library catalogues or databases.

Example 1:  Roma or Gypsies or Romanies?

Example 2:  Refugee children or Child refugees?

Example 3:  International Committee of the Red Cross or Red Cross or ICRC?

The aim of a thesaurus, then, is to ensure uniform and consistent access to information collections.  Two controlled vocabulary lists are referenced regularly in this guide: Library of Congress Subject Headings (LCSH) and the International Thesaurus of Refugee Terminology (ITRT).[13]  The advantage of searching in the subject field is that users are provided with a set of records that describe related materials.  The disadvantage is that controlled vocabulary lists are slow to incorporate new terms.  These are more likely to appear in the title and abstract fields.  If you do not know the controlled subject term, and you do not have access to the thesaurus used by the system you are searching, you can always begin with a search of all fields (a keyword search), and once you find a title that is of interest, see what subject terms are used and then search on those.

Keyword/free-text field:  As seen above, title, author and subject “keyword” searches look for the occurrence of terms anywhere within each of these fields.  On a larger scale, when the keyword field is used, the system will search for a match across fields, or anywhere within a given record, whether it be a catalogue entry, an article abstract, or even a full-text document.  For this reason, some systems label this type of search as the “free-text” option. 

Full-text Searching

Full-text search engines are generally designed to search the complete texts of the web pages or documents they index.  In theory, any word—no matter where it appears—can be retrievable via a full-text search engine.  This opens up more possibilities for a match being made than a bibliographic record simply because there are more terms available to be matched.  Also, full-text search engines allow users to bypass the sometimes rigid or old-fashioned controlled vocabulary of a subject field.  At the same time, the sheer volume of information in full-text systems like the Internet means that many more search results than a user can possibly benefit from will be retrieved.  The trade-off is greater recall but less precision.  Thus search strategies are just as important, if not more so, when searching the web or a full-text system.  The most important rule of thumb is to search on more than one term.  Studies of search habits regularly show that most users only search on one keyword.  In the web environment particularly, this strategy will always produce an overwhelming number of hits or matches.  To narrow or limit your search, either enter more search terms or take advantage of additional features provided by an individual search engine (e.g., limit by date, language, information type, etc.).  If you end up with too few hits or matches, you can broaden or expand your results by checking your spelling for errors, removing a term if you searched on several, and checking to make sure you have not entered Boolean operators in a system that does not accept them.  If the results are still not meaningful to you, consult the help section.  Search engines offer many features that can help you retrieve more relevant results, but they will vary from system to system.