Mobile Edition
Print
Get the Mag
Weekly UpdatesReflecting on their experience, the researchers suggested that "scholarly knowledge is developed through alternating phases of induction (theory building) and deduction (theory testing). When empirical observations do not confirm a theory, investigators should embark on a new phase of theory-building so that they can revise or reject the inadequate framework and replace it with a new framework." (12) Qualitative research enabled the researchers in this study to create and test a revised model that better reflected the relationship between customer service and convenience store sales.
The third way in which qualitative research can help researchers to develop valuable theories is through offering data that, if collected and analyzed appropriately, are potentially rich and powerful. Critics who do not understand qualitative research will often say that "it has an n of 1," meaning that qualitative studies represent a small sample that cannot be generalized to a wider population. However, by using frequency counts, qualitative findings can be coded into categorical variables that are analyzed and generalized using statistical techniques, such as log-linear analysis, for example (to determine the degree to which categorical groupings are associated with one another), or analysis of variance (to compare differences in categorical variables by an independent variable). Qualitative data can be used to create the grouping dependent variables used in discriminant analysis or logistic regression. Some recent organization-research analyzes qualitative findings in exactly these ways. (13)
Finally, the fourth and perhaps most important way qualitative research helps to create useful theories is by teaching researchers the realities of life in the hospitality and restaurant industries. Nothing is more useful to understanding a phenomenon being studied than to be unobtrusively "hanging around" a site, interviewing employees and guests, and observing operations. Many researchers have spent their entire careers in academe, meaning that they pursued a master's and doctorate degree following the receipt of their undergraduate degree, and along the way, obtained limited work experience--often in the form of consulting. Thus, while many well-trained researchers understand organizational problems in theory, they have little direct experience on which to draw.
Through being in the field collecting qualitative data, researchers can gain close to first-hand experience about a problem that is critical to the way they understand the phenomenon, shape their research model, and contribute to both the academic field and industry. While data-collection methods such as surveys and archival reports (e.g., annual reports, financials, press releases) are important to answering particular types of research questions, documents of that kind often allow the researcher to stay safely away from actual hospitality operations. By definition, qualitative research makes the researcher get into the trenches. It is an important training tool and represents one way for researchers to create useful new knowledge.
The Process of Qualitative Research
The process of qualitative research is guided by the researchers' purpose for the study (whether exploratory or explanatory), framing research questions, and methodology. While a discussion of the various methodologies is beyond the scope of this article, they can include the following approaches: grounded-theory analysis, deconstruction, discourse analysis, critical-theory analysis, case study, dramatism, narrative analysis, ethnography, feminist-based analysis, action research, and insider-outsider research. (14) These methodologies are based on philosophical research traditions that frame researchers' assumptions about objectivity and their own biases. Such methodologies guide the ways in which data are collected and analyzed. Yet they are all based on the primary assumption that qualitative research is interpretive inquiry. (15)
Collecting Qualitative Data
To obtain field interpretations, researchers employ a variety of qualitative methods, most of which fall under the rubric of either observation or interviews. Because the purpose of qualitative research is not to predict but to conduct a "search for understanding," (16) its design is based on researchers' spending enough time with participants that die researchers begin to form interpretations of how a phenomenon is thought about in the field. Based on the researchers' analysis--and reanalysis--of such interpretations, they develop a theoretical model.
Collecting qualitative data requires researchers to immerse themselves in the field. Participant observation involves systematically describing events and interactions among people at the site. This method is a key way for researchers to "hear, see, and begin to experience reality as the participants do." (17) Researchers trained in this form of data collection take extensive, unobtrusive, coded notes, which they immediately (within 24 hours) expand into detailed data descriptions that form the basis of a coding scheme. To ensure that the methodology has rigor, researchers must consider not only the influence of their presence on the respondents' behavior, but the way in which their own frameworks and biases may screen what types of data they see, hear, and collect.
In-depth interviewing, the other primary form of data collection, involves conducting structured r semi-structured discussions with participants. Usually researchers will prepare a set of interview questions to guide their conversations. Depending on the goals of their study, researchers may closely follow the interview protocol or they may use their protocol merely as a guide and encourage participants to expand on their responses to a question. Researchers trained in this form of data collection often tape-record their conversations with participants and, as soon as possible, transcribe their interviews into a format for data analysis. As an alternative, researchers incorporate open-ended questions into written surveys. As a way to demonstrate the complexity of conducting a simple interview, researchers must consider the following issues: the varied language and culture of all participants; whom to sample (e.g., experts, outliers, groups, individuals); how to present oneself, establish rapport, and gain tru st; and how to ensure that the ethical rights of respondents are protected, including the right to informed consent, confidentiality, and protection from any harm.
Exhibit 1 summarizes the types of qualitatively based research purposes, questions, and methodologies and various forms of data collection. Note that within the framework of observation and interviewing lie many complex forms of data collection.
Qualitative-data Analysis
The goal of qualitative data analysis is to move from summarizing the data to identifying related themes and patterns, to discovering relationships among the themes and patterns, and to developing explanations for these relationships. This iterative, non-linear data-analysis process is complex, arduous, and often messy. Once researchers write out or transcribe their data, they first attempt to make sense of it (in a time-consuming process called content analysis). Depending on the pace of conversation, for example, one 30-minute interview can generate over 20 pages of transcribed data. It would not be unusual for a researcher to have hundreds and even thousands of pages of collected data after a series of interviews.
Coding. To appropriately interpret their data, researchers categorize their notes and transcripts, creating a coding scheme to organize concepts. Usually those categories are iteratively reviewed many times. To correctly reflect the data, some are collapsed into a common category, while others are discarded as inappropriate. The categories are also juxtaposed against one another to ensure that they are conceptually distinct. The categories that result are used to create overarching themes that guide the development of a theoretical model, in which researchers present a "logical chain of evidence." (18) Qualitative software packages help code data and create categories of concepts, but the process can also be completed manually. Once researchers form their categories, they present counts that reflect the percentage or number of respondents who provided answers reflective of those groupings. Using various statistical analyses, the categories and their respective counts can be used to test hypotheses.
As mentioned, analyzing data and creating themes that accurately reflect and juxtapose all respondents' perspectives can be complex. Researchers spend many hours looking back and forth between their data and the categories they created from this data, to ensure that they are correctly reporting their findings and restraining their own biases. Often a researcher has other analysts review the data and create a coding scheme with which to compare and resolve discrepancies. Researchers also review a portion of their findings with the study's participants to check for accuracy of interpretations. The goal of data analysis is to "accumulate knowledge... through the formation of ever-more-informed and sophisticated constructions."19
Validity and Accuracy Checks
While many researchers are skilled at qualitative research, unfortunately some do perform it without adequate training. As study participants, managers will want to make sure that data are collected and analyzed appropriately--in other words, that they have "interpretative validity." (20) Some of the questions to ask researchers both at the outset and completion of the study include the following:
(1) To establish the study's credibility: How accurate or "truthful" are the findings of this study? What criteria can we use to judge them?
FEED