Some thoughts about the literature review in grounded theory studies

In a conventional quantitative study, the aim of the literature review mainly is to refine the research question, determine gaps in earlier research and identify a suitable design, and data collection method for a planned study. In qualitative research the literature search—when and how—is of a more ambiguous character. Grounded theory (GT), one qualitative method among many others, is described as a “general inductive method possessed by no discipline or theoretical perspective or data type” (Glaser, 2005, p. 141). In a GT-study, concepts are generated from empirical data rather than from existing literature. Like a detective who strives to explain what is actually happening, the GT-researcher strives to explain the main concern of participants in a specific situation/area and to find out how they resolve or process this main concern. The emerging result is presented either as a hypothesis, a model or as an abstract conceptual theory. The theory is built up around a core category and related categories. In Glaser's words, the aim of GT is to “generate a theory that accounts for a pattern of behavior which is relevant and significant for those involved” (Glaser, 1978, p. 93). Conceptualisation is a core process in GT, which thereby is a theory-generating rather than a descriptive method. Generating theory demands creative and conceptual thinking. Barney Glaser, the originator of the classical GT methodology, has stressed the importance of that a GT-researcher avoids preconceptions and remains open-minded to what actually appears in the research field. He encourages GT-researchers to “just get on and do it.” However, when a hypothesis, model or theory can be discerned in the data, a relevant literature search should be conducted and interwoven into the emerging theory. Glaser argues that an early reading of the literature (i.e., before conducting the study) is problematic. This includes that the researcher is encouraged to ignore the existing literature before entering the research field. This approach rests on the opinion that what is important in the research area will show itself repeatedly, or in other words, what is important will emerge without the “neutral” researcher is doing nothing but listen and look with an open mind. In order to understand the participants’ viewpoint, the researcher must put aside his/her personal perspective and, of course, have knowledge and competence in how to conceptualise data. Unfortunately, many researchers lack competence in conceptualisation. Any researcher has acquired considerable knowledge in the professional and disciplinary literature. To think conceptually requires that the researcher continually follows the cross-disciplinary literature, i.e., they are reading a lot. It is not easy for the researcher to put this knowledge aside when starting a new study but the point is, as I see it, not to be consciously directed by earlier theories and concepts in interpretations and conclusions of the data. One way to stay open and do good GT-studies is to maintain theoretical sensitivity through constant comparisons (e.g., constantly comparing incidents to incidents, incidents to concepts, and concept to concept) and continuous memo writing. There is a fine line between avoiding the use of literature before a study begins and being informed so that a study is focused enough. In my opinion, it is necessary to conduct an early literature review to find out if the planned study, or something similar as the planned study, has been published before. This literature review may also give a background to and motivate the interest for the particular research area. Such a literature review will also be requested by authorities when researchers apply for research grants and/or ethical permission to conduct a study or when a presumptive doctoral student applies for acceptance as a PhD candidate. At least a presumptive researcher has to demonstrate that a problem worthy of research really exists and that he/she has the necessary skill to conduct such a study. This is in line with the view of Glaser (1998) when advocating GT-researchers to do some preliminary reading before the study begins in order to put the study into a context.