Differences Between Qualitative and Quantitative Approaches to Data Analysis

Inductive and Deductive Analysis: Differences

Thick Description in Text Analysis

The concept of thick description, advanced by Gilbert Ryle and popularized in the fields of sociology and anthropology by Clifford Geertz, is used to characterize the process of putting particular emphasis on the contextual details surrounding a phenomenon when conducting qualitative research. A thick description of the phenomenon takes into account the experiential and contextual understandings of those behaviors that render a phenomenon meaningful. It involves looking at the rich details surrounding a phenomenon, and sorting out the complex details, including self-feelings, emotions, and social relationships, such that "the voices, feelings, actions, and meanings of interacting individuals are heard" (Denzin, as cited in Ponterotto, 2006, p. 540). The narration below presents a thick description of the experiences of a nurse out to improve her facility's performance in the provision of end of life care, particularly in the area of pain management (adopted from the National Academic Press, 2001).

The nurse was interested in end of life care, and devoted her time, effort, and attention to seeing it advanced in the best possible way. In her view, the facility was not doing a good job in the area of pain management and palliative care, and it was all because the degree of interaction between patients and healthcare providers was inefficient. She explains how nurses therein avoided being assigned to critical patients who needed substantial attention and care because they after all, would not be compensated for the extra effort. However, despite the numerous indications of inefficiency, physicians still did not consider pain management an issue; in fact, they found it insulting and degrading that a lower-ranking care provider thought that they were not doing what they ought to be doing.

She sought the physicians' permission to establish a pain steering committee that would improve the facility's pain management policies; they gave a go-ahead only to get her off their back, but kept frustrating the committee's efforts. Members began to withdraw from the committee once it became apparent that it was not making progress. The few who remained proposed the installation of pain scales in patients' rooms as a way of increasing the level of patient-care provider interaction and collaboration; but physicians still resisted the move on grounds that bringing patients' pain levels to their attention would cause more pain. The committee, however, remained resilient, putting in place algorithms that required nurses to graph the pain levels of their patients, assess the acceptable level of pain, and talk with them, explaining what the pain levels mean. This way, both the patient and the nurse learnt more about pain, and more about how to manage it. For a period of one and a half years, the committee developed pain resource nurses and posted them to various units to collect data and improve pain management. The nurse recognizes the danger that lies in care providers not collaborating with each other; and she aims at continuing her efforts until the facility's pain management techniques are so effective that those opposed to her efforts have no reason not to offer their support.

Part Two: Inductive and Deductive Approaches to Data Analysis

The main difference between deductive and inductive approaches to research is that whereas inductive analysis works from data to theories and concepts; deductive analysis begins with theory, and uses data to test the theory's implications (Altinay & Paraskevas, 2009). A deductive approach, also referred to as the theory-driven approach, will often begin with a hypothesis, such that the researcher moves from a general level towards one that is more specific; reading what others have done, drawing compelling theories and then testing the hypotheses resulting from those theories as shown below;

Hypothesize/theorize Analyze data hypotheses supported or not Deductive researchers develop their code schemes from theory; since such studies seek to validate or falsify existing theories, they develop the response categories from theory. The ability to draw codes from existing theory is one of the reasons why deductive approaches are important to research studies; it enables the researcher to group data according to clearly-defined codes derived from the research framework, and hence, to explain and describe relationship and interaction patterns better (Altinay & Paraskevas, 2009). However, deductive approaches are more common with quantitative research studies which are based on hypotheses testing and causality, and not on providing an understanding on why phenomena are the way they are.