All 6 entries tagged Mixedmethods
No other Warwick Blogs use the tag Mixedmethods on entries | View entries tagged Mixedmethods at Technorati | There are no images tagged Mixedmethods on this blog
March 14, 2016
Mixed Methods: Post Positivism Is No Longer Considered Appropriate
Description of and arguments against Post Positivism
Post positivism is now no longer among the set of philosophies considered appropriate for my Mixed Methods research due to my stance against philosophies that advocate pure quantitative or qualitative approaches to exploring social reality within educational contexts.
Simply put, post positivism is an extension of positivism; that it still adheres to the main concepts and principles of Positivism but modifies them at the ontological and epistemological levels but mirrors positivism at the methodological level. This modification of the concepts of positivism enables post postivism to accommodate a level of uncertainly, subjectivity, complexity and human experiences therefore recognising that absolute and certain truth about reality is not achievable. Giddings and Grant (2007) called Post Postivism a “lite” version of positivism, stating that the “post” prefix indicates a development or extension of positivism, and offer various examples of the way in which Post Positivism extends the concepts of positivism.
Positivism perceives reality as objective and independent of the mind but post positivism (along with other middle ground Philosophies) suggest that reality is embedded in its own social and cultural contexts and therefore researcher objectivity is impossible to attain. Another key area of divergence is theory verification: positivism emphasises hypothesis testing and theory experimentation in order to prove or disprove them whereas PostPositivists emphasises supporting evidence as a probability rather than being used as an absolute proof. These are just a couple of examples of where positivism and post positivism diverge at the ontological and epistemological levels. However, where they both converge and therefore enables the view of post positivism as being an extension of positivism is that it shares the same methodological assumptions.
Onwuegbuzie et al (2009) (along with many other researchers) confirms this methodological mirroring. Extent of fallibility and defeasibility of absolute knowledge accommodated by post positivism makes inferential statistics usable and applicable through inferential statistics, which utalises probabilistic approaches such as P Vales and Confidence Levels to understand reality. Post positivism also utalises qualitative data, hence post positivists can use Mixed Methods, but they use quantitative approaches to analyse qualitative data. As an example, content analysis is utilised to quantify thematic occurrences through frequency rates, and qualitative data is used in a way that enables the development of more effective quantitative approaches.
In all, post positivism is not a suitable Philosophical perspective for my Mixed Methods research because I am taking the stance that post positivism is not suited to exploring social phenomena and social reality, because everything to do with the social is too chaotic and dynamic to be represented and explained statistically. Post positivism also does not allow for much room in terms of theory building, and theory building or theorising is an aim of my Mixed Methods research as I attempt to theorise the social structures and aspects of reality that influences the phenomenon of interest. I like much of post positivism at the ontological and epistemological levels, but its mirroring of positivism at the methodological level makes it inappropriate for my Mixed Methods research. More discussions shall be found in later blog posts and more especially in my thesis.
So then: the Big Three!
With post positivism no longer being considered appropriate, this now leaves three middle ground philosophies that might be appropriate for my Mixed Methods research: complexity theory, pragmatism and critical realism. From what I have read of these so far, I have issues with pragmatism in that it appears to detach itself from philosophical and methodological concerns and places itself upon the research question. That is, the research question is the most important consideration within pragmatism and therefore all that must be done and used to answer that research question must be carried out. This has left pragmatism open to arguments that suggests it basically allows a free for all design approach with a “what works” attitude that has been questioned by a lot of writers, and I am inclined to agree with the concerns. More on this in future blog posts.
Critical realism and complexity theory appear to be the most attractive middle ground philosophies at the moment as I as yet cannot find any fault with them when it comes to exploring social reality, social phenomena, and assumptions made at the philosophical and methodological levels. Essentially, from what I can currently understand, critical realism does not concern itself with reality as a single, accessible, measurable layer (positivism / post positivism) nor does it concern itself exclusively with human experiences (interpretivism / constructivism) but it concerns itself with the underlying structures and mechanisms that produces what is found at the measurable layer and with human experiences. Now if I have interpreted this correctly, and I appreciate that what I have defined is probably a little lacking in substance but remember I am still learning and exploring this, then this makes critical realism highly applicable for substantial exploration of the social reality. Structures and mechanisms of social reality and their influence on what occurs within this social reality are highly complex and interrelated therefore complexity theory could also play a part in this structural mess.
I do perceive social reality and explorations of social reality to be highly complex and extremely uncertain, and the key to understanding the phenomenon of interest is to consider those underlying structures and mechanisms instead of constantly exploring just what is observable.
Fun stuff isn’t it? It was all a bit scary when I first started exploring Mixed Methods at this level but the more I explore the Philosophy of Mixed Methods the more interesting I find it! Lots to read and think about!
Giddings, L.S., Grant, B.M (2007): A Trojan Horse For Positism? A Critique Of Mixed Methods Research, Advances in Nurse Science, 30 (1), 52 – 60
Onwuegbuzie, A.J., Johnson, R.B., Collins, K.M.T. (2009): Call For Mixed Analysis: A Philosophical Framework For Combining Qualitative And Quantitative Approaches, International Journal Of Multiple Research Approaches, 3, 114 – 139
March 01, 2016
The Philosophy of Mixed Methods: Getting Clearer!
Things have progressed since the previous post!
Recently I have been exploring six different Philosophical perspectives that after an initial round of reading thought were most appropriate for my Mixed Methods research. Most of these advocate a middle ground approach to understanding reality that aligns with Mixed Methods methodology, and these have been Complexity Theory, Post Structuralism, Post Modernism, Post Positivism, Pragmatism and Critical Realism.
After the previous round of reading, I have concluded that there are four Philosophical perspectives that strongly advocate a middle ground approach, or in other words advocate a multiple reality perspective, that aligns strongly with a Mixed Methods methodology and they are Complexity Theory, Post Postivitism, Pragmatism and Critical Realism. Each of these shall be explored and discussed on here in time but it suffices to say here that they have the common characteristic of rejecting the Absolutism and Relativism paradigms, the opposite sides of the paradigm continuum. They reject the idea that reality can be understood either through Absolutism or Relativism, and therefore place emphasis on the view of reality as a mixture of observable, measurable, deterministic and controllable elements, and also elements that are dynamic, chaotic, unobservable, and cannot be reduced to variables. This leans suitably towards a Mixed Methods methodology, but the extent to which each paradigm advocates Mixed Methods differ, and the writers and practitioners within each paradigm differ further the extent to which they advocate Mixed Methods methodology. The most common paradigm used is Pragmatism but just because a paradigm is more dominant it doesn’t mean that it is most suitable for my own research.
The other two paradigms Post Structuralism and Post Modernism are not suitable as a guide of Mixed Methods inquiry but are suitable in building a platform upon which Mixed Methods can be criticised. Both paradigms reject the modernist perspectives of reality (e.g., postivism, absolutism, and so on) and strongly advocate a multiple reality perspective therefore lean fairly strongly towards relativism and constructivism paradigms. From the readings that I have carried out so far, both perspectives appear to criticise Mixed Methods on Philosophical grounds: that Mixed Methods orientate towards Positivism, that there is a series lack of Mixed Methods researchers engaging at a Philosophical level, and therefore that there are various ontological and epistemological issues that remain unresolved within a Mixed Methods context. So whilst they do not reject Mixed Methods outright as an interesting and useful methodology, Post Structuralists and Post Modernists criticise Mixed Methods methodology at the Philosophical level and therefore have been critical of the Pragmatist approach.
So there we are! Four paradigms that have shown promise as a guide of Mixed Methods inquiry, and a couple of paradigms that are not useful as a guide of Mixed Methods inquiry but useful in understanding the criticisms and critiques of Mixed Methods methodology. It’s alright in any thesis to write loads about the wonderfulness of a methodology but the criticisms need equal attention and solutions need to be developed, explained, applied, and evaluated, all of which I aim to achieve in my thesis.
January 17, 2016
Methodology now in place: the convergent flavour of the Triangulated Mixed Methods methodology!
The methodology has been set in place and that is the Mixed Methods methodology; specifically, Triangulated Mixed Methods methodology (triangulation simply means to collect and analyse data from multiple sources using multiple methods in order to increase validity and reliability of the research findings: more about this shall be discussed in time). There are various flavours of Triangulated Mixed Methods each of which having a specific, clear, concise and contextually defined set of objectives therefore each flavour is suitable for a particular purpose. Out of all of these flavours I have decided to select the convergent flavour of the triangulated mixed methods methodologies.
Triangulated Mixed Methods Methodology: the Convergent flavour.
This convergence design has been termed in various ways in existing literature including “convergent parallel” design, but regardless the aim of this flavour is to converge quantitative and qualitative findings at the interpretation level. This shall enable the findings to be compared, contrasted, corroborated and related (hence convergence) in order to discover similarities and differences in order to increase validity and reliability of research findings (hence triangulation). But there are other interesting potential uses for this converged (or mixed) results such as developing further research methods to explore further aspects of the phenomenon that were not been previously considered.
Other varieties of Mixed Methods and indeed other flavours of the Triangulated mixed methods differ in the order of which quantitative and qualitative data should be collected and analysed, whether or not the quantitative or qualitative data should be independently collected and analysed or integrated at various stages, and the importance or weighting of both types of data. The Convergence model encourages the separate, independent collection and analysis of quantitative and qualitative data, but converge at the interpretation stage. The model also promotes a concurrent (where both quantitative and qualitative can be collected simultaneously within the same time phase of the research) approach to collecting data instead of sequential (quantitative then qualitative or qualitative then quantitative). This means that the findings of the quantitative does not influence the findings of the qualitative, and vice versa. Basically they are both collected at the same time, but do not influence each other or data collected using other methods. What this can do however is influence the design of any further methods that might be used throughout the duration of the research. Further to this, the model also encourages an equal weighting of quantitative and qualitative data in answering research questions and dealing with aspects of the research problem.
The key decisions have been made with each decision bringing about more questions and challenges that need to be addressed, but this is the case with all decisions made about research design. If you are not generating any questions about your research design as you go along then your inquiry into your own thinking, perspectives about reality, purposes and uses of your own design and a complete and full understanding of the underlying problems and questions and the relationship between these and the design shall be unguided and chaotic. Questions bring order and a sense of direction to any research project, that their development and refinement are continuous, and is something that each Ph.D. candidate should be engaged with at all levels and stages of their Ph.D. research.
Therefore, the selection of the Mixed Methods methodology, the selection of the type Triangulated Mixed Methods, and the selection of the Convergent flavour, along with the previous selections of specific methods that shall collect quantitative and qualitative data and the ongoing decision making regarding data analytical methods, introduces many more challenges and questions than answers! There simply does not appear to be any right or wrong answer or approach to deal with any of these challenges or questions: what therefore needs to be done is attain a full understanding of each challenge and question, carefully read and analyse relevant literature, and develop a solution or answer with suitable argumentation.
Key questions are: could mixing of the data occur at both the analysis and interpretation stages? Would this approach to mixing be appropriate for my research? What could the potential findings be? What implications could this have on any aspect of the research design? What implications could this have on the rigour, validity, reliability, generalisability, completeness and comprehensiveness of the findings, discussions, and research design overall?
I am not in a position to answer these questions yet, but they along with all other questions and challenges, and every other question that shall occur in the future, shall be answered in time!
‘till next time: let your research design be guided not by your answers but by your questions!
January 15, 2016
Revisiting Philosophical and Methodological problems, and solving them!
During the past few blog posts I have been discussing the various Philosophical and Methodological problems that have occurred even before the time that Constructivist Grounded Theory was initially considered as an overall methodology. It was realised that this really was not going to work and therefore scrapped plans to use Constructivist Grounded Theory as a methodology and instead decided to use Mixed Methods methodology; more specifically, Triangulated Mixed Methods methodology, whilst using Constructivist Grounded Theory as a method along with other methods to collect and analyse data. It did not take long to realise that this form of Mixed Methods would practically answer those questions that I have been asking for a while.
The central problem I had was reconciling differing Philosophical perspectives: Positivism and Interpretivism. Even when Constructivist Grounded Theory was replaced with Triangulated Mixed Methods methodology the reconciliation between widely differing Philosophical perspectives was not apparent straight away.
Whilst reading through the Philosophical literature of Mixed Methods methodology it became apparent that merging or combining these perspectives is impossible and not desirable in Mixed Methods methodology. This is because the differing Philosophical perspectives entail particular methodologies which themselves entail the development of particular methods that address different but related questions, sub questions and problem areas of a wider research project. In my Ph.D. research for example the research problem and research questions have been designed in a way that entails the need for differing Philosophical perspectives and therefore differing methodologies and methods to properly investigate the phenomenon of investigation.
From this, Mixed Methods methodology makes sense to use because it enables the mode of inquiry to encompass two related but separate Philosophical perspectives that deals with different questions and problems of the same phenomenon. But what of the Philosophical perspective of Mixed Methods? There is a general consensus in the literature that the Philosophical perspective of Mixed Methods is Pragmatism, which is a Philosophical orientation of reality that defines research design in terms of its usefulness and that it works to an acceptable level. What this basically suggests is regardless of the Methodological perspectives that are a part of a research design, as long as it is purposeful and explores the phenomenon of investigation as required, then it’s an acceptable research design. In addition to Pragmatism some papers reference methodological relativism as the key perspective but this shall need further exploration. Some initial thinking suggests that Pragmatism and Relativism along with Contextualism and Interpretivism could be all related in some way but this needs further exploration. There appears to be plenty of debate and discussion of Philosophical perspectives of Mixed Methods methodology and these shall be explored in time.
In all however, Pragmatism appears to be the overarching Philosophical perspective and has therefore answered my question about Philosophical reconciliation, but shall be reading further into Pragmatism as a research design Philosophy. So, what of the methodological problems?
Any Philosophical problem of a research design introduces a Methodological problem because they are related. The Methodological problem that was encountered at the time therefore was the merging or combining of quantitative data and qualitative data in order to correlate, combine, merge, and corroborate data as fully as possible in order to present a holistic analysis of the phenomenon of investigation. Just like the Philosophical problem, it was realised that the Methodological problems would be resolved simply through introducing Mixed Methods methodology.
Mixed Methods methodology offers a number of different solutions to the problems of mixing quantitative and qualitative data and I have decided to go for mixing at the interpretation stage. What this means is both quantitative and qualitative data shall be analysed separately using their respective sets of analysis methods, but the discussion shall corroborate, compare, contrast, and relate quantitative findings and qualitative findings. This obviously has been noted in academic literature as being challenging and I shall talk about this another time, but in the meantime I think this approach to dealing with the methodological problem works. Additionally, mixing could take place at the analysis level therefore qualifying the quantitative data and quantifying the qualitative data although I am not sure at this time if this would be really relevant. This shall need further investigation and thinking.
So in summary: the previous Philosophical and Methodological problems have been resolved simply through changing the methodology itself, but this has introduced pages of new questions that need exploration, and the answers to those questions shall produce more questions but that is the way research works!
‘till next time: you won’t learn unless you ask, but be prepared to ask more questions from the answers!
January 09, 2016
Methodological Breakthrough, Part B: Introducing The New Methodology!
Introducing Triangulated Mixed Methods Methodology
Ta da! This has come as a breakthrough for my research as I have now identified what I believe to be the research methodology that is most suitable for my research. Triangulated Mixed Methods is a research methodology that applies Triangulation approaches within the context of Mixed Methods research, which essentially according to some writers enable higher levels of validity and reliability through comparisons and corroborations of differing types of data from different sources, which exactly matches my vision of my research project.
From the initial rereading, Cresswell provided the clearest and most useful definitions of this type of Mixed Methods methodology that convinced me of its suitability. Cresswell describes Triangulated Mixed Methods Methodology as suitable for research projects involving comparisons, validations and expanding discussions between quantitative and qualitative findings. This is suitable for my project because it will involve comparing quantitative data with qualitative data and using these further analytical comparisons and discussions to expand on separate analyses and discussions that shall be made with each data set in the thesis.
There are other reasons, but that was the major, influential definition of Mixed Methods that has encouraged the favoured methodological view to Triangulated Mixed Methods.
What does all this mean now for my research Methodology and research Philosophy?
In brief: Triangulated Mixed Methods methodology is now the research methodology for my Ph.D. with Constructivist Grounded Theory now being used as a research method along with questionnaires. Interview and focus groups shall be used in addition at a later stage as and when deemed necessary. This obvious impact on my methodology will have an impact on my research Philosophy, although the Philosophical assumptions and perspectives of Triangulated Mixed Methods, and Mixed Methods in general, appears to be highly discussed and debated by a lot of authors and Philosophers (oh fun!)
So will this methodology make reality any easier to understand?
Er, no, well, it will, eventually! Basically, even years before starting the Ph.D. I had an idea that my research would be quite complex because what I am doing is exploring perceived learning (quantitative data, qualitative data) and actual learning processes that take place (qualitative data, mostly). This direction has not changed; it has only became more specified and detailed but I am not going to discuss the specifics on here: I shall leave them to my future published research papers and thesis. The methodology now selected makes a lot more sense to me because it allows me to investigate the phenomenon in exactly the way that I envisioned.
Loads. Sheer absolute loads to do, which is fine because it gives me plenty of blog material! Methodologically speaking, I need to select the most appropriate variant of the Triangulated Mixed Methods methodology to use, as there are several variants that have been designed and debated, although I already have a fair idea but need to do more reading and experimenting into this. Also, I need to identify Philosophical assumptions and develop Philosophical arguments for using Mixed Methods methodology and this shall take a little while given the amount of debates from various authors. Following this, I then need to carefully plan the way that Constructivist Grounded Theory and Questionnaires shall work effectively within a Triangulated Mixed Methods methodology, and carefully think about the practical assumptions and considerations that Triangulation makes upon the data analysis. Not only this, but I also need to carefully consider the Philosophical assumptions, arguments, practical applications and so on of both Constructivist Grounded Theory and Questionnaires and the way that a Triangulated Mixed Methods methodology actually bring these Philosophical and Methodological differences together in the way that research objectives are achieved.
Additionally I need to carefully consider the way in which the methodology and methods all come together to deal with issues of data validation, integrity, reliability, consistency, coherence, authenticity, and so on, and also develop ways in which challenges that each method and the methodology provides shall be carefully managed, maintained and dealt with so that any data errors are avoided as best as possible.
All this and much more shall be considered within the thesis and various research papers that shall be published from the research. Now that the methodology and methods are set, I can begin to think about, within the context of my research, all these Philosophical, Methodological and practical issues and much more than has been discussed here as I think I have wrote enough about the subject for the time being!
‘till next time: is there really such a thing as objective reality?
October 11, 2015
Grounded Theory? Mixed Methods? Both? Current ponderings!
In a blog post yesterday I suggested that my research was going to be based on a Grounded Theory approach using Mixed Methods methodology. After spending till after midnight yesterday reading up on the subject and from what I have been reading this evening I can safely assume that I was talking complete nonsense. Well, possibly, but that's the beauty of learning: you think about things, and you develop your ideas and approaches based on your continuous learning and thinking, and the beauty of having a blog such as this is that the thinking, learning and development of ideas can be documented! So, where is my thinking at the moment? Methodologically speaking I can push the research in a couple of different ways: use a pure Grounded Theory approach, or used a Mixed Methods approach using Grounded Theory to explore qualitative data. That’s where my current thinking is: I know for sure that other types such an Ethnographic study, a Phenomenological study, an Action Research study, a Narrative Research study and so on are not appropriate for the aims of this research, and that is to develop a new theory that explains the relationship between constructs of a phenomenon of investigation.
I could use Grounded Theory and use quantitative and qualitative data to generate a new theory that explains this relationship among constructs, or I could develop a Mixed Methods study that initially uses Grounded Theory to develop a theory from qualitative data, and then collect and analyse quantitative data to experiment with this new theory. But I’m not sure at this time if I want to actually implement such a research design because the goal is to create a new theory and not to create and then experiment with the new theory: I’m not sure that there is enough time to create such a massive study and then having to write about everything that there is unless I am allowed to hand in a thesis that is a couple of hundred thousand words!
I’m quite happy that I’m really thinking about this because it shows engagement not just with the phenomenon of investigation but with the research methodology and the extent to which I can develop and push research methodologies and methods to really explore and acquire knowledge of that phenomenon. This refers to what I could class as the Philosophy of Research Methods and thinking about the research methods at this level includes the following questions in relation to my own research:
What extent could Mixed Methods methodology with Grounded Theory explore reality and the relationship between the constructs of the phenomenon that is being investigated?
What impact would the different Mixed Methods approaches have on the findings, and therefore on the development of the theory?
Comparing a pure Grounded Theory approach to a Mixed Methods approach, which method could really assist with contributing towards theoretical development?
What extent could Grounded Theory be pushed to explore the behaviour of a phenomenon of investigation?
Could Grounded Theory represent a more authentic reality and therefore provide the basis of a more convincing explanation of reality than Mixed Methods? Or vice versa?
The current contention is, I shall be using multiple methods within either methodology (Grounded Theory and Mixed Methods are known as methodologies although different authors have classed differently but they all have essentially the same meaning) so determining the answers to the questions asked might in part be answered by the way in which each methodology handles the multiple methods that I shall be using.
So, a lot of thinking to do! I shall be exploring both Mixed Methods and Grounded Theory methodologies over the next few weeks to really find out which would be most suitable for the aim of my research and the questions that I want to explore.
This shall also prove to be interesting for the Upgrade process because in the report and presentation a line of reasoning needs to be given as to why a certain methodology has been selected over other methodologies. Therefore, comparing Grounded Theory with Mixed Methods methodologies in the context of my research should provide plenty of material to work with!