All entries for Friday 25 May 2018

May 25, 2018

Some Latest Thoughts on the Literature Reviews and Research Design

I have lost count of the amount of times I have rewritten literature review sections and I am now completely rethinking the structure, layout, content and even the number of literature review chapters given the planned changes to the research design. I have quite frankly given up on all ideas of being settled on any kind of literature review format, layout etc. till the day I actually print out the thesis……..

I have a lot of thoughts about the literature reviews. I am now planning on going for at least two literature review chapters with the first engaging with the relationship between society, culture, education and other concepts that I have now come to know as relevant, and the second focussing on the concepts and characteristics of the phenomenon of interest and technologically enhanced learning. The third chapter was going to focus on the exploration and examination of the different theories and models used to explore the phenomenon of interest in different ways. I am not sure now though whether it is best to keep the planned third chapter the way it is, or to discuss and critique existing models and theories when I have developed my own model of what it is I am exploring. However because my research design is emergent, there is a requirement to save critiquing empirical literature most relevant to the phenomenon of interest till the later chapters where literature can be integrated with research findings in order to compare findings, and to authenticate and validate the emerging model or theory. As you can tell I’m not yet decided about the third chapter because of the complexity of the research design……….

Thoughts On The Research Design

The research design has changed because I have now come to fully realise the multi-dimensional and multi-layered nature of the phenomenon of interest. More importantly, I have come to realise or have become more aware of the characteristics of the data that I want to explore as a result of further rereads of the data and, therefore, the result of coming to realise that Grounded Theory simply isn’t going to capture these characteristics.

This realisation has led me to viewing the phenomenon of interest as multi-layered and, therefore, the need to carry out a multi-level approach to data analysis. There is care needed here with language: there is a difference between a multi-level approach and a multi-staged or multi-phased approach. I am saying that the phenomenon of interest can be explored using different levels; in the case of my research, three levels, but I need not go into any detailed explanations as to what they are on here at this time. Therefore I am saying that the phenomenon can be understood in three different ways, but combined they can provide potentially a powerful insight into the complexity and process of the particular learning phenomenon of interest. Whether or not the multi-level perspective of the phenomenon of interest shall lead to a multi-staged (e.g., Mixed Methods) research design remains to be determined. It is likely though to become mixed methods with the way I am currently thinking about the way I would like to investigate the phenomenon of interest.

What I can say is the first phase naturally aligns with the work I have already completed: Grounded Theory coding, or at least the first stage Open Coding. I am not, however, sure at this time if I need to fully develop the codes into grounded theory categories, or if they can simply be left as they are and not call it Open Coding but simply call it another coding process. Either way, I shall be rereading the data again and reread all the codes and theoretical notes that I have made, and the product of the rereading and reanalysis of the data should lead me to decide exactly what further analytical methods I shall be adopting in the further phases.

I do like the idea of using the graph / network analysis as previously discussed and I feel that there is a need for some sort of quantitative analysis of the data (which would make it mixed methods) but I need to ground this need in the data and the literature (though I’ve read enough to consider these approaches as possible).

Grounding the need to change or amend a research design within an emergent research context is an important point to make, because it is easy to think (as has been argued by some authors) of emergent research designs as ‘anything goes’ but this isn’t the case. Not every research project, particularly emergent designs, is fully planned at the beginning stages of the research project. It can take some time and several reanalysis of the data for the design to really emerge and this accompanies the way in which the researcher becomes sensitised or becomes aware of the extent and complexity of the phenomenon of interest, and the way in which is the best approach to understanding this complexity.

What’s important in my current thinking is not that you are able to perceive multi-level complexity or that you potentially or eventually come to the idea that you need to combine various methods in order to capture this complexity, but that you can fully and elaborately justify your choices and justify why you perceive the phenomenon of interest in the way that you do. Everything has to be grounded in data as well as in sound, authentic reasoning and logic that can stand up to scrutiny (which goes right up to your ontological and possibly meta-philosophical considerations), and of course in the literature.

The next step now is to move away from writing the literature review for a while and refocus on analysing the data and continuing to draft the methodology chapters, but I shall explain this further in the next blog post that shall be coming soon. The coming summer months shall be spent therefore mostly on data analysis!

‘till next time!