Margaret Olson. Encyclopedia of Case Study Research. Editor: Albert J Mills, Gabrielle Durepos, Elden Wiebe. Sage Publications. 2010.

Documents, as a record of human activity, provide a valuable source of data in case study research. Along with interviews and observations, they comprise one of the main forms of data sources for interpretation and analysis in case study research. Document analysis can provide a window into a variety of historical, political, social, economic, and personal dimensions of the case beyond the immediacy of interviews and observations.

Conceptual Overview and Discussion

Documents can include a wide variety of materials relevant to the case under study. In Sharan Merriam’s view, documents comprise a variety of written, visual, and physical material, including any artifacts that can shed light on the case. In this view, documents are by-products of human activity that “document” their activity over time. Because these documents have been created through ongoing day-to-day activities unrelated to the present research they can provide authentic records that shed light on multiple facets of the case and can provide clues that may be useful in framing interview questions and observation protocols. Documents can be categorized into three main types: (1) public records, (2) personal documents, and (3) physical materials. A fourth category of documents are those created by the researcher, such as field notes and a research journal. Many of these documents may be created and/or stored in an electronic environment.

Although at times there is a clear distinction between public and private documents, at other times the line between them is quite blurred, creating a continuum from completely public, to semi-public, to semi-private, to completely private. Personal documents can now be found anywhere along this continuum. With burgeoning online venues such as personal Web pages, Facebook, and numerous blogs, the personal easily becomes available to the public. Traditional public records include anything that is available for public use, although some of these are more readily available to the public than others. These documents include (but are not limited to) census data; newspaper articles and archives, maps; court documents; wills; company and government policy documents; manuals; handbooks; photographs; magazines; books; brochures; and advertisements, from billboards, to flyers, to television commercials. The myriad range of public documents available requires creativity on the part of the researcher in identifying which might be most relevant to the particular case. Along with researcher creativity, research participants can point to valuable documents the researcher might not have thought of, especially those that are not completely public. Thus, whereas documents can help lead to good interview questions, interviews can also lead to new document sources. This category could include semi-public documents, such as employee contracts, manuals and policy handbooks, minutes, payroll lists, health records, and so on, to leads on personal documents, such as photo collections, family histories, diaries, letters, scrapbooks, videos, and other electronically stored family or individual records. This category of documents, by representing accounts created by individuals or particular interest groups or organizations, can provide valuable perspectives on the case that likely will not be found in documents originally created for the general public. Finally, artifacts and physical traces as documentation of human activity can provide rich data sources to analyze.

Document analysis begins as documents are identified and/or selected on the basis of their usefulness or relevance as data for the particular research. A parallel step is to validate the source and authenticity of the chosen documents. Knowing the source of documents enables the researcher to determine their likely purpose or intent as well as determine the perspective from which they were created. Merriam suggested several questions researchers might ask about the authenticity of a document:

• What is the history of the document?
• Is the document complete, as originally constructed?
• If the document is genuine, under what circumstances and for what purpose was it produced?
• What were the maker’s sources of information? Does the document represent an eyewitness account, a secondhand account, a reconstruction of an event long prior to the writing, an interpretation?
• Do other documents exist that might shed additional light on the same story, event, project, program, context? If so, are they available, accessible? Who holds them?

Merriam comments on the importance of distinguishing between primary and secondary documents. Primary documents tend to be those created by people closest to the phenomenon under study. Secondary documents are those created by those not directly involved and perhaps at a later date. It is possible for a document to be either a primary or a secondary document depending on the purpose of the research. For example, a photographer’s pictures could be primary documents in a study focused on the photographer but secondary documents if the objects or people in a photograph were the primary research focus. Once the selected documents are located and authenticated, they need to be copied and/or photographed. These copies of original documents can then be sorted into appropriate categories, and further analysis can be carried out.

Document analysis in terms of who has ready access to or who created which documents in a case can lead to insights about relationships among case participants, power structures, and communication patterns. Documents can be analyzed in terms of demographic and historical information to provide background to a case. They can be compared for confirmatory or contradictory information that can illuminate new understandings or further questions to pursue. Content analysis, which has been a staple of document analysis, can take the form of highly structured searches that count the number of times particular words or phrases are used. Jonathan Hull described evaluation processes for computer programs designed to perform various document analysis functions on large volumes of data such as checks. The constant-comparative form of document analysis described by Robert Bogdan and Sari Biklen is a much more fluid search for themes or meanings that recur in a variety of documents leading to categories. Depending on the research focus, researchers may be searching for descriptive information, may do a comparative analysis of different perspectives from different sources, or may do a critical analysis that would uncover hidden sources of power.

Application

Document analysis is a helpful research tool that can be used in as wide a variety of ways as a researcher has the imagination to create. A few examples are suggested here. A variety of documents, including such items as letters, sketch-books, original artwork, and critics’ reviews, could provide the total data source in a case study of the evolution of an individual artist or a particular art movement. A combination of public documents, such as newspaper accounts and public notices, could be compared with a variety of personal journals, photographs, and letters to tease out some of the contradictions and complexities of a particular case. Kelly Asmussen and John Creswell, in studying perceptions of campus safety after a gunman attack on campus, discovered an increase in newspaper advertisements for and registration in self-defense classes. Through records kept by the campus bookstore they identified increased sales of mace and whistles, and a sharp increase in cell phone purchases, especially by women, was noted. These documented changes, when compared with those before the gunman incident, pointed to a dramatic increase in safety concerns. A document such as a map or photograph could be used to stimulate interview responses. A high school social studies teacher had students compare copies of high school yearbooks from 1948 up to the present day to examine shifts in the school community’s gender and ethnic composition, shifts in the kinds of extracurricular activities, and shifts in various awards given, which can mirror changes in the surrounding community. A critical analysis of promotional bro-chures for an institution or business can lead to the identification of belief systems that may lie below the surface of espoused purposes.

Critical Summary

Because they were originally created outside of the scope of the research, documents can provide a rich source of data pertaining to the day-to-day activities surrounding the case. Documents can also provide historical information that lead to a better understanding of the case in question and can provide a diverse wealth of information from a wide variety of sources. This wealth of possibilities creates several issues of which researchers need to be aware. Researchers need to be cognizant that because different documents were created for different purposes, they will present different points of view. To limit bias, researchers need to have a wide enough variety of documentary sources to provide a reliable research report. On the other hand, much of the documentary data may be irrelevant to the present research purpose, and the sheer volume of documentary data can become overwhelming. Thus, researchers using documents as a data source need to both broaden and narrow their scope, ensuring that they use documents that provide a rich variety of information focused on their particular research purpose.