Secondary Sources in Social Research

Secondary Sources in Social Research "The most fundamental drawback [of relying on secondary sources] stems from the fact that this previous research is likely to have been done with different aims… It may also have been based on assumptions, and even prejudices, which are not readily discernible, or which are inconsistent with those of the current research…" (Blaikie, Norman, 2009, p. 161). Conducting research into social issues nearly always requires the use of secondary sources. That's just a fact.

No doubt using primary sources in social research is vitally important as well, but for most researchers -- students, scholars, journalists, authors and psychologists among others -- using secondary sources is the meat and potatoes of their work. They rely on the research of others simply because primary sources are not always readily available. Meantime, what are the issues that challenge researchers when using secondary sources? That question is approached critically in this paper, using the published scholarly (peer-reviewed) literature and other reliable sources.

Defining Secondary Sources -- A Quick Review Author and sociologist Chris Livesey points out that secondary sources are those that "already exist" and have been researched and written down by someone "other than the researcher currently using it" (Livesey, 2006, p. 2). The practical reasons for using secondary sources -- which are newspaper articles, books, scholarly peer-reviewed articles, magazines, television, radio, and films -- include the savings of time (that it would take to dig up the primary sources from which the secondary sources were acquired), money, and certainly effort, Livesey explains.

There are also several methodological reasons for utilizing secondary sources: a) in historical research, such as "Centuries of Childhood" by Aires, the author dug into original paintings and documents to "support his idea that childhood was a relatively recent invention"; b) in comparative research, data such as Durkheim used in "Suicide: A Study in Sociology" is available to the researcher, saving him or her weeks of digging into suicide statistics from a number of countries. Issues Associated with Secondary Sources in Social Research Authors Richard M.

Grinnell and Yvonne Unrau -- writing in their book, Social Work Research and Evaluation: Foundations of Evidence-Based Practice -- clearly have serious qualms about secondary sources. In fact they are very picky even when approaching primary sources, suggesting that good solid research should go into choosing primary sources. "Did the writers have a vested interest in the outcome of an event?" they ask (Grinnell, et al., 2010, p. 407).

"Were their beliefs involved in the event?" The authors continue, adding, "We must always assess the value of the primary sources before we decide which ones to use" (p. 407). Given that this book warns writers and researchers to always question the veracity of primary sources, one can be certain the authors are extremely skeptical about secondary sources. "Secondary sources are accounts, including books and articles, that are based on the analyses of primary sources," they explain.

And although secondary sources "are important for framing questions and getting a context for an investigation," the authors point out, "with few exceptions historical reports" that are based solely on secondary sources "…add little that is new to our knowledge of the past" (p. 407). Moreover, they continue, by using secondary sources "exclusively" the researcher / writer is taking a chance that he or she may "perpetuate the errors made by earlier writers" (p. 407). Grinnell points out a classic example of perpetuating errors when using secondary sources in socially-based narrative on page 408.

In this instance, an author wrote a descriptive account of a riot in an African country in the 1930s. The number of deaths that the author of this account reported was way off the mark because he "…chose to rely on local newspapers alone, and the newspapers of colonial Africa printed only a portion of the truth" (Grinnell, p. 408).

The account of the riot that emerged in this author's work "…was obviously biased," Grinnell continues; it reduced a "complex human event to a recitation of dry half-truths, all properly footnoted" (p. 408). Grinnell points out another example of the potential folly of using secondary data exclusively when covering social issues on page 408. In the late 19th Century, agents from the U.S. Government were estimating the number of American Indians in a particular Indian reservation.

"Some agents reported more American Indians than the actually were present" in order to beef up the supplies that were to be allocated by the United States Office of Indian Affairs, Grinnell explains. The problem was, the resulting population numbers for those particular tribes "can be accepted only with caution"; and albeit the U.S. agent in this matter was a primary source, the carelessness with which the data was complied makes the resulting Indian population skewed for future researchers (Grinnell, p. 408).

While questioning the use of secondary sources is good advice, for most students and other lay researchers, the ability to obtain the primary documents from which their secondary sources were gleaned is often out of the question.

Furthermore, while Grinnell and Unrau are being very professional, and very editorially cautious about sources, when a researcher is using a peer-reviewed, scholarly paper from an established publisher in a respected journal, by quoting and paraphrasing carefully, and referencing through competent citations and bibliography, the researcher is being scholastically responsible and is not in jeopardy of plagiarism or reckless presentation. If there are errors in the peer-reviewed materials being used, it is not the responsibility of the researcher / writer using those materials to make the revisions. Meanwhile, Dr.

Emma Smith, head of the Department of Education and Social Justice at the University of Birmingham, UK, presents the "Pitfalls and Promises" of using secondary data in educational research. Although Smith understands that there is justification for skepticism when approaching secondary data, she also believes that secondary data "offers numerous methodological, theoretical and pedagogical benefits" (Smith, 2008, p. 323). In her research Smith finds that educational research is "widely viewed as having an 'awful' reputation" in both the U.S.

And England, so she invested time and energy into conducting research into the use of secondary data vis-a-vis educational issues in the UK. A report in 2004 in England (conducted by the Economic and Social Research Council, ESRC) posited that the lack of "quantitative skills is endemic in many areas of Social Science" (Smith, p. 326). Hence, the ESRC concluded, "…there is an urgent need to enhance research quality" (p. 326).

Smith dug into "well-regarded journals" in the fields of Sociology, Social Work, and Education that had been published over a seven-year period of time. What did Smith find in her research? About one quarter of all the scholarly journals she investigated "adopted some form of quantitative method" (492 journals out of a total of 2016) (p. 326). And of those 492 journals using quantitative formats, 41% utilized secondary data analysis (that is 202 out of 492), Smith explains.

However, looking closer at the data, about 10% overall of all the papers that were reviewed used "secondary data" but the majority of Sociology papers (which is just one of three categories) used secondary data (that is 75% of Sociology journals), according to Smith's research. Less than one half of all Education journals (42%) that used numeric methods "involved the analysis of secondary data"; and among the Educational journals that did employ numeric methods of research, the "vast majority made use of school performance data" (Smith, p. 327).

Interestingly, while a substantial number of British researchers in Education "report using secondary data analysis," Smith asserts, "…very few actually appear to use the technique in their published research" (p. 328). She sees this lack of usage by Education scholars as "further evidence… for a lack of methodological pluralism in this part of the field" (p. 328). Having introduced the data on the number of social science and educational journals in England that use secondary data, Smith points out the "potential pitfalls" of using secondary data for analysis vis-a-vis social issues.

Yes, she agrees, there are many promising uses of secondary data, and in many instances the data can be "analyzed and replicated from different perspectives," providing opportunities for the discovery of relationships that were not given consideration in the primary research. However, she continues, the pitfalls are there and they have probably contributed "most" to the dramatic "under-use of secondary data analysis" and also have contributed to the "exaggerated suspicion of social measurement & #8230; and excessive distrust of the officially produced numeric data" within the community of social scientists (p. 328).

To wit, the very nature of secondary data "leaves it particularly susceptible to criticism," Smith writes; indeed the research used often embraces "the analysis of data that has been collected with a very different purpose in mind" (p. 328).

An example of this is the British Social Attitudes Survey; in this survey -- which is based on interviews with citizens -- the secondary data analyst making the interviews "…is far removed from the source of the data and may be unaware of, or unconcerned with, the context in which the research took place" (p. 328).

Smith boils it down to two main concerns (pitfalls) and objections to the use of secondary data in social research: a) "It is full of errors"; and b) "…because of the socially constructed nature of social data, the act of reducing it to a simple numeric form cannot fully encapsulate its complexity" (p. 328). The author offers the national Census as a perfect example of.