In the field of qualitative research, a very famous method is "grounded theory" (1967) proposed by Glaser and Strauss. Grounded theory is a qualitative approach to research, the main aim of which is to build a theory from empirical data (Strauss, 1987: 5). The researcher usually begins the study without theoretical assumptions, and starts directly from actual observations, making empirical generalizations from the primary data, and then moving on to theory. This is a method of building a substantive theory from the bottom up, i.e., searching for core concepts reflecting social phenomena on the basis of systematic data collection, and then constructing relevant social theories through the connections between these concepts. Rooted theory must be supported by empirical evidence, but its main characteristic is not its empirical nature, but the fact that it abstracts new concepts and ideas from empirical facts. In terms of philosophical thought, the Zaganian theoretical approach is based on a post-positivist paradigm that emphasizes the falsification of already constructed theories.
The rooted theory approach originated in a field observation by both Glass and Strauss (1965, 1968) in the 1960s in a hospital on the handling of dying patients by the medical staff. The development of this aspect is related to two theoretical ideas, coming from philosophy and sociology: one is American pragmatism, especially the ideas of Dewey, G. Mead, and Peirce, who emphasized the importance of action and focused on working with problematic situations and generating methodology in problem solving; the other influence came from the Chicago School of Sociology, which made extensive use of field observations and in-depth interviews data collection, emphasizing the understanding of social interactions, social processes, and social change from the perspective of the actors.
I. The basic idea of rootedness theory
The basic idea of rootedness theory mainly includes the following aspects.
1. Producing theories from information
The Zagan theory places special emphasis on upgrading theories from information, believing that only through in-depth analysis of information can a theoretical framework be gradually formed. It is a process of induction, whereby information is continuously condensed from the bottom up. Unlike general grandiose theories, rooted theory does not logically deduce the researcher's own pre-set assumptions, but rather inductively analyzes the information to begin with. Theories must be traceable to the original sources from which they are derived and must be based on empirical facts. This is because rooted theorists believe that only theories derived from information are viable. If the theory matches the information, the theory has a practical use and can be used to guide people's concrete life practices.
The primary task of rooted theory is to establish substantive theories (i.e., theories applicable to a particular space-time) between grand theories and micro-operational hypotheses, but it does not exclude the construction of formal theories that are universal. However, formal theories must be built on top of substantive theories, and formal theories can be built on top of all kinds of related substantive theories only after substantive theories have been built on top of information. This is because, according to rootedness theory, knowledge is cumulative, a continuous process of evolution from facts to substantive theories and then to formal theories. Constructing formal theories requires a large number of sources and the mediation of substantive theories. If a formal theory is constructed directly from a single source, the jumps involved are too great, and many gaps are likely to be created. In addition, formal theories do not have to have only a single form of composition, but can cover many different substantive theories, integrating, condensing, and generating many different concepts and perspectives into a whole. This kind of intensive formal theory is richer than those single formal theories, and can provide meaningful explanations for a much wider field of phenomena.
2. Sensitivity to Theory
Since the main purpose of Rooted Theory is to construct theory, it places special emphasis on the researcher's sensitivity to theory. Whether in the design stage , or in the collection and analysis of data, the researcher should be sensitive to their own existing theories, previous theories and theories presented in the data, and pay attention to capture new clues for constructing theories. Maintaining theoretical sensitivity not only helps us to have a certain focus and direction when collecting data, but also to pay attention to looking for concepts that can express the content of the data in a more concentrated and condensed way when analyzing the data, especially when the content of the data itself is relatively loose.
Often, qualitative researchers are more adept at detailed descriptive analysis of the phenomenon under study, and are not particularly sensitive to, or interested in, theory building. Rooted theory, out of its own particular concern that theory has greater explanatory power than pure description, emphasizes the researcher's sensitivity to theory.
3. The method of constant comparison
The main analytical idea of Zagan's theory is comparison, which is to compare and contrast information and theories, and then extract the relevant categories and their attributes according to the correlation between the information and theories. Comparison usually involves four steps: 1) Compare the data according to conceptual categories: after coding the data and assigning them to as many conceptual categories as possible, compare the coded data across the same and different conceptual categories, and find attributes for each conceptual category. 2) Integrate the relevant conceptual categories with their attributes, and compare these conceptual categories, considering the relationships that exist between them. (2) Integrate the conceptual genera with their attributes, compare the conceptual genera, consider the relationships between them, and link the relationships in some way. (3) Outline a preliminary theory, determine the meaning and scope of the theory, and return the preliminary theory to the original source for verification, while continuously optimizing the existing theory to make it more refined. (4) Present the theory, describing the information, the conceptual genera, the attributes of the genera, and the relationships between the conceptual genera, layer by layer, as an answer to the research question. The answer to the research question is to describe the information available, the conceptual categories, the properties of the categories, and the relationships between the conceptual categories.
4. Theoretical Sampling Methods
When analyzing the data, the researcher can use theories initially generated from the data as the criteria for sampling the data in the next step. These theories can guide the next steps in data collection and analysis, such as selecting data, setting codes, and establishing coding and filing systems. Each of the theories presented at the moment has a guiding role for the researcher and can qualify where and how the researcher should go next. Therefore, data analysis should not just stop at mechanical linguistic coding but theoretical coding. The researcher should continually build hypotheses about the content of the data, generate theories through round-robin comparisons between the data and the hypotheses, and then use these theories to code the data.
5. Flexible Use of Literature
Using relevant literature can broaden our horizons and provide new conceptual and theoretical frameworks for the analysis of data, but at the same time, we should be careful not to use theories of our predecessors too much. Otherwise, the ideas of the former may bind our thinking, so that we intentionally or unintentionally other people's theories to their own data set, or in other words, their own data to other people's theories set, that is, what people say, "cut the feet to fit the shoes," rather than "tailor-made "
This is the first time I've seen this.
While using previous theories appropriately, rootedness theory suggests that the researcher's personal interpretations can also play an important role in constructing the theory. The researcher can "understand" the source material because the researcher brings in his or her own empirical knowledge, and the theory generated from the source material is actually the result of the continuous interaction and integration between the source material and the researcher's personal interpretation. There is a triangular interaction between the source material, the researcher's prior understanding, and the results of previous research, and the researcher must combine the source material with his or her own personal judgment when using the literature. The researcher himself should develop the habit of questioning himself and being questioned, listening to the multiple voices in the text, and understanding the interaction between himself and the primary sources and the literature.
6. Theoretical Evaluation
The Rootedness Theory has its own criteria for checking and evaluating the theory, which can be summarized as follows: 1) Concepts must be derived from the primary sources, and after the theory has been built up it should be possible to go back to the primary sources at any time, and a rich content of the sources can be found to serve as the basis of the argumentation. 2) Concepts in the theory should be sufficiently developed, i.e., the density within the theory should be relatively high, i.e., the concepts within the theory should be well developed. The density should be relatively large, i.e., there are many complex concepts and their meaning relations within the theory, and these concepts are situated in a dense theoretical context. In contrast to what Geertz (1973) calls "deep description", which emphasizes conceptual density, "deep description" is mainly a dense picture of the phenomenon at the descriptive level.3 ) Each concept in the theory should have a systematic relationship with other concepts, and the concepts themselves should be fully developed. Each concept in a theory should be systematically connected with other concepts, "theory is a rational connection between concepts and sets of concepts" (Strauss & Corbin, 1994: 278), and the concepts should be closely interwoven to form a unified and intrinsically connected whole. (4) The theory linked by the set of concepts should have strong application value, should be applicable to a broader scope, with strong explanatory power, theoretical sensitivity to the subtleties of the behavior of the parties, and can raise theoretical issues related to these phenomena.
Second, the operation procedure of rooted theory
The operation procedure of rooted theory generally includes: 1) generating concepts from information, and logging in the information step by step: 2) constantly comparing the information and the concepts, and systematically asking generative theoretical questions related to the concepts; 3) developing the theoretical concepts, and establishing the connection between the concepts and the concepts; 4) theoretical sampling, and coding the information systematically; 5) constructing the theoretical concepts; 6) developing the theoretical concepts, and establishing the connection between the concepts and the concepts. coding of information; and 5) constructing theory, seeking to obtain density, variability, and a high degree of integration of theoretical concepts. Level-by-level coding of information is the most important aspect of rooted theory, which includes three levels of coding.
1. Level 1 coding (open access)
In level 1 coding (open access), the researcher is asked to adopt an open mind and try to "suspend" personal "bias" and "stereotypes" of the research community. In Level 1 (open coding), the researcher is asked to "suspend" personal "bias" and "stereotypes" of the research community as much as possible, and to log all the information as it is presented to the researcher. This is an operationalization process that breaks up the collected data, assigns concepts to them, and then puts them back together in a new way. The purpose of logging is to discover conceptual categories from the data, to name the categories, to determine the genus and dimension of the categories, and then to name and categorize the phenomenon under study. The process of open logging is similar to a funnel, starting with a wide range of logging and then narrowing down until the code numbers become saturated. When logging data, the researcher should ask specific, conceptually related questions about the content of the data. It is important to keep in mind the original purpose of the research while leaving room for unanticipated goals to emerge from the data. An important principle that the researcher should observe at this stage is to believe everything and believe nothing (S trauss , 1987: 29).
In order to keep their analysis going, the researcher should stop frequently to write analytical memos while open-endedly logging into the data. This is an effective means of analyzing the data and can prompt the researcher to reflect on the theoretical issues that arise in the data and to gradually deepen the preliminary theories that he or she has already constructed through writing. The main purpose of this round of logging is to open up the data to inquiry, and all interpretations are preliminary and undefined. The researcher's main concern is not so much what concepts are in the text at hand, but how it might enable deeper inquiry.
There are some basic principles to consider when conducting open-ended logging: 1) Log the data carefully, not missing any important information; log as carefully as possible until saturation; if new code numbers are found, further primary data should be collected in the next round. 2) Look for words used by the person in question, especially those that can be used as code numbers. 3) Give each code number an initial name. 4) Look for the words used by the person in question, especially those that can serve as code numbers. 5) Give each code number a name. 6) Give each code number a name. 7) Give each code number a name. 8) Give each code number a name. 3) Give each code number a preliminary name, either in the person's own words or in the researcher's own words, without worrying about the appropriateness of the name. 4) As you analyze the data line by line, ask specific questions about words, phrases, sentences, actions, meanings, and events, such as: How is this data relevant to the study? What categories can be generated from this event? What specifically does this information provide? Why did these things happen?5 ) Quickly analyze some of the dimensions of the concepts related to the information, which should evoke cases for comparison; if no cases are generated, they should be sought immediately.6 ) Note the relevant entries in the listed log-in paradigm.
2. Secondary coding (associative logging)
The main task of secondary coding (also known as associative logging or axial logging) is to discover and establish connections between conceptual categories in order to express the organic links between the various parts of the data. These connections can be causal, chronological, semantic, contextual, similar, differential, reciprocal, typological, structural, functional, process, strategic, etc. In axial coding, the researcher can find and establish various connections between conceptual categories in order to show the organic connection between various parts of the data. In axial logging, the researcher only analyzes one category at a time, looking for relevant relationships around this one category, so it is called "axial". As the analysis progresses, the connections between the various categories should become more and more specific. When analyzing the relevance of conceptual categories, the researcher should not only consider the relevance of the conceptual categories themselves, but also explore the intentions and motives of the people who express them, and put their words into the context of the time and the socio-cultural background in which they are located.
Once the relationships between each set of conceptual categories have been established, the researcher needs to distinguish between primary and secondary categories. Once these different levels of categories have been identified, the researcher can link the relationships between them by means of comparison. Once all the major and minor relationships have been established, the researcher can then reorganize the source material in new ways. In order to find out whether these ways of analyzing make sense in practice, the researcher can also build a prototype theory construct guided by an action or interaction orientation after exploring the relationships between the various categories. This theoretical construct focuses on dealing with real problems and solving them, and its theoretical basis is the practical rationality of the person concerned.
3. Three-level coding (core logging)
Three-level coding (also called core logging or selective logging) refers to the systematic analysis of all the conceptual categories found and the selection of a "core category", and the analysis is continually focused on the code numbers related to the core category. The core category must repeatedly prove to be overarching in comparison with other categories, encompassing the majority of findings within a broader theoretical context. Like the strings of a fishnet, the core genus can hold all other genera together as a whole, acting as a "catch-all". In summary, the core categories should have the following characteristics: 1) The core categories must occupy a central position in all categories, more concentrated than all other categories, with the largest number of categories between the meaning of the correlation, the most powerful to become the core of the data. 2) The core categories must appear frequently in the data, or those indicators that express this category must appear in the data with the greatest frequency; it should express the importance of a particular category in the data. 3) The core categories should have the following characteristics: 1) The core categories should have the following characteristics: 2) The core categories should have the following characteristics: 3) The core categories should have the following characteristics: 4) The core categories should have the following characteristics It should be a recurring and stable phenomenon in the data. 3) The core category should be easily associated with other categories, and these associations should not be forced, but should be established very quickly, and the associations between each other are very rich in content. (4) A core category in a substantive theory can easily be developed into a more general theory; before it can be developed into a formal theory, it needs to be carefully examined and tested in as many substantive areas as possible. (5) As the core categories are analyzed, the theory naturally moves forward. (6) As the core categories are continually logged in terms of dimensions, properties, conditions, consequences, strategies, and so on, the core categories can be used as the basis for the theory's development. The search for internal variation is a feature of rooted theory, as the core categories are constantly being logged in terms of dimensions, properties, conditions, consequences, strategies, etc. Subordinate categories can become very rich and complex.
During the core registration phase, the researcher should always ask: "At what level of generalization can this conceptual category(ies) belong to a larger category of social analysis? Is there a more important core of conceptual categories that can be generalized? How can I connect these conceptual categories into a systematic theoretical construct?" The memo written by the researcher during this period should be more focused on the density of theoretical integration of the core genera, with the aim of integrating the theory until theoretical saturation and completeness is achieved. Once the core genera have been found, it can provide direction for the next step in theory sampling and data collection.
The specific steps of core logging are: 1) to identify the story line of the data; 2) to describe the primary and secondary genera and their attributes and dimensions; 3) to test the initial hypotheses that have already been established and to fill in the conceptual genera that need to be supplemented or developed; 4) to pick out the core conceptual genera; and 5) to establish systematic links between the core genera and the other genera. If we find more than one core category at the beginning of the analysis, we can connect the related categories by constant comparison and eliminate those that are not closely related enough.
Let me give you an example of the three-level coding process. In my study of the intercultural interpersonal activities of some Chinese students studying in the U.S. and the interpretation of their meanings (1998), I logged the data in a hierarchical manner. First, in the open log, I found many "local concepts" used by the interviewees, such as "interest, desire, have come and gone, be prepared, often, in-depth, care for others, take care of others, manage, save face, lose face, be subtle, be considerate, be tolerant, do things in an official way, communicate emotionally, be enthusiastic, be warm, be a good friend, be close, be a good friend, be a good friend, be a good friend, and so on". warm, warmth, buddy, closeness, reward, stray, circle, unsettled, insecure, overwhelmed, big kid, inferior, national pride, uncomfortable, etc.". Then, in the associative register, I found some connections between the above concepts by linking them under seven main categories: "interaction, human, emotional exchange, making friends, outsider, self-esteem, change". Under each of the main categories there are related subcategories, such as "caring for and taking care of others", "understanding and tolerance", "saving face" and "being subtle" under "favors"; and "outsiders" under "outsiders", "outsiders", "outsiders" and "outsiders". Under "outsider", there are "being out of the loop, circle, overwhelmed, unstable, insecure, lonely, homesick, free and at ease", and so on. Finally, after all the categories and relationships were established, I identified the core category in the core logging process as "cultural constructions of the self and human-ego relationships". After further analysis of the primary data within this theoretical framework, I developed two rooted theories: 1) culture has an orienting effect on an individual's concept of self and self and interpersonal behaviors; and 2) cross-cultural interpersonal interactions have the function of reconstructing an individual's self-cultural identity.
4. An Example of Analysis
In the following, let me borrow an example provided by Strauss in Qualitative Analysis for Social Scientists (1987: 12-7) to demonstrate the process of analyzing data by the Zaganian theoretical approach.
The hypothetical research question is: "Does (and how does) the use of machines and equipment in hospitals affect the interaction between medical staff and patients?" We see a lot of machine devices attached to patients in the wards, and we can form an initial category - "machine-body connections" - to - to represent this phenomenon. Based on our observations, we tentatively decided to divide the machines into two main categories: those connected to the outside of the patient's body (e.g., the skin) and those connected to the inside of the patient's body (e.g., the nose, the mouth). This distinction leads to two dimensions of the genus "machine-body": internal connections and external connections. These dimensions can then be further subdivided, for example, in relation to 'internal connections', we can continue to ask: 'Do these machines cause pain to the patient? Are they safe for the patient? Are they comfortable? Are they scary?" When asking these questions, we can use a dichotomy: "yes" or "no", or we can use a continuum from "strong" to "weak". or a continuum from "strong" to "weak". Of course this categorization comes not only from the raw data we collect in the field, but also from our own empirical data (e.g., these internal organs of a human being are so sensitive that a machine connection may cause pain in these areas; that pipe sticking out of a patient's stomach looks scary, so the pipe may be unsafe).
The question above relates to the consequences of the behavior or thing: "If this thing looks like this, then might it have life-threatening consequences? "At this point, we can add some other specific conditions, such as: if the patient moves too fast, or if he rolls over in his sleep at night, or if this tube falls out and his body becomes inflamed, in which case his life would be endangered. We can also ask questions about the strategies used by the medical staff, "Why do they keep the tube in this way and not that way?" Or ask about the strategy used by the patient, "Did he negotiate with the nurse to use a different approach?" We can also ask about the interaction between the parties, "What happened between him and the nurse when the machine was attached to him? Did they tell him in advance, did they give him some warning? Did they do it without telling him, and as a result he was horrified?" (This last question also relates to the outcome that resulted from the interaction between the parties.)
After these questions have been given initial answers, we can begin to form some hypotheses. Some of these hypotheses will need to be further tested by observation or interview, but we can do so in a more focused way. We might find that a tube connected to a patient's nose is uncomfortable but safe. So we can interview about that. If we want to continue to inquire about "conditions that make it unsafe," we can ask the nurse, "At what point are these connections unsafe for the patient?" We could also look for conditions that occur when the patient's nose becomes unsafe for the machine to connect to, such as a sudden break in the connection or a problem with the way the connection is made.
This thread of thought can further lead us to subdivide the dimensions, ask more questions, and form more initial hypotheses. For example, for those machine connections that are more prone to dislodgement, we might ask, "How did they dislodge? Was it because of an accident, negligence, or was it intentional (e.g., the patient pulled it off herself when she was annoyed, uncomfortable, or scared)? What strategies and techniques did the nurse use to avoid or prevent dislodgement as much as possible? Giving special care ? Warn the patient not to move? Emphasizing that personal safety depends on not moving or pulling off the connection no matter how much it hurts ? Or by 'working together' to ensure that the connection is only made for a few hours? Or periodically removing the machine so they can relax?" These questions, assumptions and distinctions may not be "true", but if they are, we can explore them further and find "yes-no-maybe" and "yes-no-maybe" and "yes-no-maybe". -Possibly" and "Why". Obviously, we always end up asking more questions about the conditions and consequences, not only for the patient, but also for the patient's relatives, the nurses, the different staff, the functioning of the ward, and possibly the redesign of certain parts of the machine.
The more focused inquiry above leads us naturally to ask, "Where can I find evidence for 'X' or 'Y'?" This question raises the issue of "theoretical sampling". Through the previous survey, we have begun to look for people, events, and actions to sample for our emerging (and perhaps primitive) theories. For the novice researcher, this sampling is usually done covertly in the context of a comparative exercise that focuses on contrasting different sub-dimensions.
Guided by the above theory, we can also sample more broadly. For example, sampling the safety and comfort of other machines to see if they are connected to the human body, such as X-ray equipment, airplanes, toasters, hoeing machines, or the shock to the body when the machine vibrates in the hands of those hired to break up the concrete pavement in the street. This comparison is not intended to form a generalized theory of all machines or safe/hazardous machines, but rather to provide theoretical sensitivity to theories about the use of medical equipment in a hospital setting. Our external sampling is closely linked to our internal sampling. Of course, these comparisons can also be drawn from other sources of our own experience (so-called "anecdotal comparisons"), such as our own personal experiences with machines, watching others use them, and our own reading of novels, autobiographies, or reportage about machines.
This paper focuses on the methodology of rooted theory in qualitative research, especially its basic ideas and operational procedures. There are many other different ways of constructing theory in qualitative research, and individual researchers may take a different approach to and deal with theory depending on the genre in which they are trained, the way they look at the problem, and the context in which they are researching. Therefore, readers need to adopt an open and flexible attitude when considering the issue of theory construction.
Chen Xiangming (1998): Sojourners and "Foreigners"--A Study of Chinese Students' Intercultural Interpersonal Interaction, Changsha: Hunan Education Press.
Geertz, C. (Ed.). (1973).The Interpretation of Cultures. NewYork: Basic Books.
Glaser,B.(1978).Theoretical Sensitivity. Mill Valleyociology Press.
Glaser,B.(1982).Generating Formal Theory. In R. G.Burgess (Ed.), Field Research: A So urce Book and Field Manual.London:George Allen & Unwin ( Publishers) Ltd.
Glaser,B.& Strauss,A.(1965).Awareness of Dying. Chicago:Aldine.
Glaser,B.& Strauss,A.(1967). The Discovery of GroundedTheorytrategies for Qualitative Research. Chicago: Aldine.
Glaser,B.& Strauss,A.(1968). Time for Dying. Chicago:Aldine. Strauss, A. (1987). Qualita tive Analysis for SocialScientists. Cambridge: Cambridge University Press. Strauss, A. & Cor bin, J. (1990). Research: GroundedTheory Procedures and Techniques, Newbury Park: Sage.