Grounded Theory is most accurately described as a research method in which the theory is developed from the data, rather than the other way around. That makes this is an inductive approach, meaning that it moves from the specific to the more general. The method of study is essentially based on three elements: concepts, categories and propositions, or what was originally called �hypotheses�. However, concepts are the key elements of analysis since the theory is developed from the conceptualization of data, rather than the actual data.
Strauss &
Grounded theory is often perceived as a method which separates theory and data but others insist that the method actually combines the two. Data collection, analysis and theory formulation are undeniably connected in a reciprocal sense, and the grounded theory approach incorporates explicit procedures to guide this. This is especially evident in that according to grounded theory, the processes of asking questions and making comparisons are specifically detailed to inform and guide analysis and to facilitate theorizing process. For example, it is specifically stated that the research questions must be open and general rather than formed as specific hypotheses, and that the emergent theory should account for a phenomenon that is relevant to participants.
There are three distinct yet overlapping processes of analysis involved in grounded theory from which sampling procedures are typically derived. These are: open coding, axial coding and selective coding. Open coding is based on the concept of data being �cracked open� as a means of identifying relevant categories. Axial coding is most often used when categories are in an advanced stage of development; and selective coding is used when the "core category", or central category that correlates all other categories in the theory, is identified and related to other categories.
Data collection is directed by theoretical sampling, which means that the sampling is based on theoretically relevant constructs. Many experiments, in their early stages, use the open sampling methods of identifying individuals, objects or documents. This is so that the data�s relevance to the research question can be assessed early on, before too much time and money has been invested. In later phases, a systematic relational or variational sampling is frequently employed with the objective of locating data that either confirms the relationships between categories, or limits their applicability. The final phase generally involves discriminate sampling, which consists of the deliberate and directed selection of individuals, objects or documents to verify the core category and the theory as a whole, as well as to compensate for other, less developed categories. Also included as necessary parts of the analysis are other procedures such as memo writing and the use of diagrams, as well as procedures for identifying and incorporating interaction and process.
Grounded theory contains many unique characteristics that are designed to maintain the "groundedness" of the approach. Data collection and analysis are consciously combined, and initial data analysis is used to shape continuing data collection. This is supposed to provide the researcher with opportunities to increase the "density" and "saturation" of recurring categories, as well as to assist in providing follow-up procedures in regards to unanticipated results. Interlacing data collection and analysis in this manner is also designed to increase insights and clarify the parameters of the emerging theory. At the same time, the method supports the actions of initial data collection and preliminary analyses before attempting to incorporate previous research literature. This is supposed to guarantee that the analysis is based in the data and that pre-existing constructs do not influence the analysis and/or the subsequent formation of the theory. If existing theoretical constructs are utilized, they must be justified in the data.
Grounded theory provides detailed and systematic procedures for data collection, analysis and theorizing, but it is also concerned with the quality of emergent theory. Strauss & Corbin state that there are four primary requirements for judging a good grounded theory: 1) It should fit the phenomenon, provided it has been carefully derived from diverse data and is adherent to the common reality of the area; 2) It should provide understanding, and be understandable; 3) Because the data is comprehensive, it should provide generality, in that the theory includes extensive variation and is abstract enough to be applicable to a wide variety of contexts; and 4) It should provide control, in the sense of stating the conditions under which the theory applies and describing a reasonable basis for action.
Grounded theory offers many advantages, however because it is such a painstakingly precise method of study, it requires high levels of both experience and acumen on the part of the researcher. For this reason, novice researchers should avoid this method of study until they have achieved the proper qualities needed to effectively implement the approach.