How to Choose and Evaluate Sources

By Jacqueline Dirks

Types of Sources

Examples of Sources

Evaluating Sources

Useful Links for Evaluating Sources

Types of Sources

You will need to critically evaluate the available sources related to your chosen topic. Secondary sources are historical works that offer interpretations of historical evidence and debates, and usually (but not always) draw on some combination of primary sources, and the interpretations of other historians, to support a particular argument. A secondary source is usually in the published form of a book or periodical article from a scholarly journal. A primary source is usually an actual or facsimile record or document from the period under study, for example, newspapers, census records, letters, photographs, or films. Primary sources, like secondary sources, need to be assessed rather than simply accepted at face value; a document from the period under study often reflects the biases of its writer or compiler. Since most secondary works of history rely on primary sources, it is logical to describe these first.

Examples of Primary Sources

A primary source is usually an actual or facsimile (e.g., photocopied) record or document from the historical epoch under study. Primary material can include sources such as unpublished papers, letters or documents left by individuals or organizations. Published works from the period under study, such as autobiographies (e.g., Progressive era reformer Jane Addams' 1910 Twenty Years At Hull-House or radical anarchist Emma Goldman's 1931 Living My Life) or journalists' accounts of an event (e.g., John Hersey's 1945 description of the aftermath of the atomic bombing in Hiroshima) may also be used to support historical arguments. Oral testimonies that have been recorded on tape and/or transcribed in some format are yet another type of primary source. Legal and political institutions also generate documents that provide historical evidence. For example, depending upon their interests, historians may utilize the transcripts of criminal or other trials or the texts of judicial court decisions. Types of official government documents include the records of state and federal census data (the first census was recorded in 1890 in Massachusetts), authorized documents published by state and federal agencies, and the minutes of Congressional proceedings published in The Congressional Record.

Many primary sources are available in libraries, some in the form of published books or preserved on microfilm or microfiche; most unpublished archival documents are housed in manuscript or special collections. While many libraries will lend books, microfilm and microfiche, it is important to note that archives do not lend documents, though some will make copies for a fee. In addition to documents available on microfilm or microfiche, some primary sources, such as government statistics on census data, have been prepared in CD-ROM or other computer-accessible forms. Depending on the source, these may be available in selected libraries or via the World Wide Web.

Many primary sources are available in published collections of reprinted documents. For example, during the economic depression of the 1930s the United States government funded the Works Projects Administration (WPA), a federal agency which employed out-of-work writers and academics to collect interviews with former slaves, immigrant and native-born workers in various trades, and pioneer settlers in different parts of the country. Other document collections bring together the works of acknowledged historical figures, such as Princeton University Press's collected papers of Woodrow Wilson or the selected works of W.E.B. DuBois edited by Eric J. Sundquist in The Oxford W.E.B. Du Bois Reader (1996). Some document collections bring to light information about people who were not famous. For example, Ira Berlin's multi-volumed set of documents from the Freedmen's Bureaus includes testimony on the post-Civil War South given by Bureau officials and the freed people (ex-slaves) who applied to them for aid.

Historians sometimes analyze works of fiction, such as novels or short stories, as historical evidence of attitudes or ideas expressed in a given time. For example, historian Jeanne Boydston compared nineteenth-century sentimental domestic fiction, in which housework was portrayed as occupying very little of women's time, with evidence of actual work done by women in the nineteenth century northeastern United States. She found that while the sentimental fiction made housework invisible, women's diaries and other sources showed that women contributed significantly to the household economy. Depending on the period and topic you wish to research, published fictional works may also be available in libraries.

Secondary Sources

Secondary historical works offer interpretations of primary material, and usually (but not always) draw on some combination of primary sources, and interpretations of other historians, to support an argument. A secondary source is usually in the published form of a book or periodical article from a scholarly journal. Though some secondary sources, mainly recently published articles, are available on-line, most books and scholarly journals still exist in some form of print medium. Some exceptions include Reviews in American History (Project Muse) which features issues from 1996 to the present on-line.

Book titles can be found in many places, including computerized library card catalogs, and bibliographies and footnotes of articles and books. Periodical articles from scholarly journals are also important sources. To find articles on your chosen topic, check the Reader's Guide to Periodical Literature, which was established in 1900. Another good source is the America History and Life index, which is available in bound issues for works published up to the 1980s. There is also a computerized version of these indexes. For more information about how to use these, go to library research.
There are many specialized historical journals, and you should always check to see if there is a relevant journal that may have sources on your topic. For example, if you are researching the role of women in African American families, you might want to check the indexes of the Journal of Negro History (founded in 1916) or the Journal of Family History (established in 1976), among others. Back issues of bound periodicals are located on library shelves-many of these also publish annual indexes that are shelved with the bound issues. Scholarly articles are often important in historical debates, since much new work is initially published in article form.

Historical journals that cover specific fields in the discipline can also be good sources of book reviews of important works in the field. Some books may be reviewed in more than one journal. Some journals are devoted exclusively to book reviews (for example, Reviews in American History and The New York Review of Books) while others include reviews of individual books or several works in a particular field. As with any other source, you need to assess book reviews critically, since they vary widely in their quality and fairness.

Computer-Accessible Sources

Some primary sources, such as government statistics on census data, have been prepared in CD-ROM or other computer-accessible forms. Remember that much of this information has been entered by someone who has transcribed rather than scanned the document. Transcription errors are potential problems in all documents that are not photographically reproduced.

Depending upon one's source and argument, historians may need to have the original document at hand. For example, the selected digitized images made from photographs and lantern slides taken by reformer Jacob Riis, located at the Web site for the Museum of the City of New York Online are quite impressive. Riis (1849?-1914) worked to abolish child labor at the turn of the century, and used lantern slides and still photographs to illustrate his lectures and books. But in order to fully understand how Riis manipulated and deployed photographic images to move his audiences, the historian needs to examine slides not in isolation, but alongside the lectures they illustrated. Riis also staged many of his photographs, bribing his subjects to hold a particular pose or surprising them by igniting photographic flash powder. In some instances one can detect the way that Riis "framed" his subjects, but only where one has a full rather than a cropped image. Photographs and line drawings were used to illustrate Riis's most famous work, the book How the Other Half Lives: Studies Among the Tenements of New York (1890). A hypertext edition of How The Other Half Lives has also been made available via the Internet, courtesy of David Phillips of Yale University. The hypertext edition reproduces the entire text and all of the illustrations (photographs and drawings) from the original print edition published by Charles Scribner's Sons. While this makes it easy for historians to quickly scan Riis's language for key words, it is important to keep in mind that people in the 1890s would have either heard Riis's words during a lecture accompanied by stereopticon slides or read the printed text.

Finally, as of 1998, only particular sources have been translated to this medium. In their article "Brave New World or Blind Alley?" historians Michael O'Malley and Roy Rosenzweig found that the content and quality of on-line history resources vary widely, in part because of who funds and/or maintains the various sites. If you rely heavily on surfing the Internet as a research strategy, you will miss many potential topics and sources and your final paper will be poorer for it.

Though some secondary sources, mainly recently published articles, are available on-line, most books and scholarly journals are still in some form of print medium. Some exceptions include Reviews in American History (Project Muse) which issues from 1996 to the present on-line.

The Internet can make it possible to communicate with archivists and glimpse the contents of archives far from Reed College. However, at this date only a limited number of (usually well-funded) archives have made readable, complete digitized documents available to on-line researchers. The Library of Congress, for example, has everything from early silent films to oral histories available at its Web sites. Many archives, from state historical societies [example] to Yale University's Manuscripts and Archives, have provided on-line guides to their holdings, specific finding aids and information for researchers. In some cases, the material must be used at the archive in question. But some archives will make copies of specific documents for distant researchers, for a fee. (The charge usually includes the costs of reproduction and mailing.) Check the Web site for information about whether the archive will photocopy or film materials; email an archivist if you need a specific document. And remember, if you cannot get a reproduction of the relevant document in the first six to eight weeks of the term, chances are good that you will not be able to use it in your Junior Seminar paper.

How to Evaluate Sources

Once you have chosen a suitable history research topic and begun your research, you will need to critically evaluate the relevant secondary and primary sources to which you gain access. Historians make arguments and offer interpretations of the past events and available evidence; you need to assess which arguments you find persuasive and articulate why one interpretation seems more plausible than another. Primary material also requires careful scrutiny. Documents do not speak for themselves-if they did, there would be no need for historians. A document from the period under study often reflects the perspective or position of its writer or compiler; primary sources, like secondary sources, need to be assessed rather than simply accepted at face value.

As you examine your sources, you will need to gather specific information about them. Always write down the complete title, publisher, place, and date of publication of any source. It is a good idea to get into the habit of jotting this information down for all titles that you think you may use-in general it is much easier and faster to check your notes than it is to repeat the entire research process in order to find a citation. This information will form your bibliographic entry.

To compose the annotation for each entry, you will need to try to critically evaluate your sources. Your main task in writing the annotation is to assess how the source might be significant for your research. You should also try to determine the reliability or merit of secondary sources. Try to assess the author's credentials and place in the debates. Is there information about whether the author (or authors) is an established scholar in the field, a graduate student, an independent scholar, a journalist? Has he or she published several or many books and/or articles on similar topics? Is his or her work cited in most of the works you have found? To answer such questions about the authors of books, check the acknowledgments. If your source is an article, check for editorial comment near author's name or at the head of the footnotes; if you have the entire issue of the journal in which the article appeared, also check back of journal for "Notes on Contributors". You can often form a general idea of where an author stands in the field or debate by reading the Introduction and Conclusion of the book in question.

Useful Links for Evaluating Sources
This link offers a list of critical questions to ask when evaluating a web site.
A lengthy article written by a college library director for presentation at Harvard. She reviews some of the organizations that rate websites, and provides her own criteria.

Photo Credits: 1937 WPA photo. FAP: Tonawanda Reservation. Sponsored by Rochester Museum: woman and man boiling bark in ashes and water for fibre.