Business research methods and applications

¶ … primary source is one in which the author has direct experience with the subject. Some examples of primary sources are court documents, texts of laws or Supreme Court decisions, photographs, videos, eyewitness accounts and other similar documents. A primary source can either be objective (a written law) or it can be subjective (an eyewitness account). In addition, a primary source could be a text -- for example The Odyssey -- while a secondary source would be a scholarly interpretation of The Odyssey. A secondary source is one in which the information is described by a person with a degree of distance from the event or action. In particular, secondary sources rely on analysis, interpretation or description. Tertiary sources are sources that compile information from secondary sources. An example of a tertiary source would be an encyclopedia.

Secondary sources are the most common and viable of the three types, which may account for why they are most used. Primary sources are sometimes unavailable in convenient format, or may be too biased to be of much use. Tertiary sources are often distant, and while they can have value in terms of overview, their best information often comes from secondary sources. This leads researchers to go directly to those secondary sources. For many management questions, secondary sources are the most feasible. Studies that measure any independent variable against firm performance, for example, are secondary sources by their nature. Many management issues are studied using proxies such as results from the annual report, again showing why secondary sources are common in management study.

3. There are a few problems associated with secondary data. The first is that secondary data inherently contains interpretations or analysis. Therefore, secondary data is not objective, but is often rather subjective. That many people who guide researchers apparently fail to grasp this only exacerbates the problem. This subjectivity is not only in the interpretation of data, but also in the design of research studies, so the researcher must take a critical eye to any secondary information from the beginning, analyzing the attitudes and philosophies that underlie the research design itself. Another problem with secondary data is that it can be difficult to acquire a balanced view of secondary data, as different sources may have different underlying philosophies but do not address the same questions. Secondary data may, in fact, not address the research question directly at all, but may rather address the issue at hand only tangentially. There may not be secondary sources that directly address the research question.

4. Qualitative research differs from quantitative research in a number of different ways. The most superficial is the use of numbers, or lack thereof. Whereas quantitative research emphasizes the study of factors that can be explicitly measured, qualitative data is more descriptive. This has a couple of key implications. The first is that qualitative data is inherently more subjective. It is directly subject to interpretation from the researcher. Quantitative data can be interpreted by the researcher, but ultimately the data is presented in raw form and can therefore be interpreted by another researcher as well. The other implication is that whereas qualitative data is interpreted at the observation level, quantitative must be analyzed at the root level. The way in which is survey is designed will reveal the biases of the researcher, so that is the level that must be given the most scrutiny. Beyond this, quantitative research often reflects the use of proxies, so there is some distance between the numbers generated and the actual phenomenon being studies. This especially true of business, where for example the number of takeovers is used as a quantitative proxy for the qualitative element of corporate governance. The researcher must be aware of the imperfections of the proxy; with qualitative data it is more the interpretation that must be subject to scrutiny than the researcher design itself.

5. Data from qualitative research is substantially different from data from quantitative research. Superficially, qualitative research consists of descriptors that are non-numeric. Quantitative research is based strictly around numeric measures. The result of this is that the two must be interpreted entirely differently. The data is also acquired in a different manner as well. Qualitative measures are often gathered through direct observation, while quantitative results can often be derived through primary sources other than observation (annual reports in business, for example). The method of gathering the data may not differ, but the degree to which the data is objective and the degree of subjective interpretation of the data will differ between qualitative and quantitative data.