ARTICLE 98: Research Methods for Ph. D. and Master’s Degree Studies: The Layout of the Thesis or Dissertation Part 2 of 9 Parts

Written by Dr. Hannes Nel

I discuss deconstruction and empirical generalisation in this article.

Is deconstruction just a euphemism for plagiarism?

After all, what we do when we deconstruct a concept, argument, knowledge or philosophy, is to take what somebody else said or wrote and change it to serve our own purpose.

You be the judge if deconstruction is theft or progression.


Deconstruction is not an independent research method as such, but rather a way in which data that you collected for your research is ‘unpacked’ into more useful chunks that belong together and that can be articulated to the purpose of your research. To rearrange the data, you need to identify the right meanings for terminology and concepts.

Constructivism as a paradigm addresses deconstruction. Some academics are of the opinion that deconstruction belongs with post-structuralism. However, it is important to also discuss it separately as part of the process of research methodology seeing that it is necessary, regardless of your paradigmatic preference.

To clarify the difference between constructivism as a paradigm and deconstruction as a research method – constructivism deals with the way in which people perceive their research environment; deconstruction deals with the way in which you, as a researcher, will contextualise and articulate the research data that you collect to convey the ‘message’ of your investigation.

When deconstructing data that you collected, you will group them under headings and sub-headings that will enable you to offer the data in harmony with the purpose of your research, hopefully on a higher level of abstraction or at least in a more creative manner. When studying towards a doctoral degree you will need to ‘create’ new data, which will probably include some deconstruction.

When doing research for a master’s degree and even more so for a doctoral degree, you will need to group your data into a set of categories and transform the groupings into abstract types of philosophies and knowledge which you need to analyse further. Dedicated computer software enables you to code your data so that deconstruction is much easier to accomplish by just grouping pieces of information under specific codes and then analysing and recombining the information into new messages. In this manner you can reconstruct the data that you collected into a logical, accurate and authentic thesis or dissertation.

Deconstructing data is not about disclosing an already established, underlying or privileged truth, thereby committing plagiarism. Rather, it is about synthesising existing data in such a manner that the inherent truth of the data is extracted and offered as an alternative, higher level construction of reality. In the case of doctoral studies such deconstruction should lead to alternative meanings, aligned with the problem statement, problem question or hypothesis of the research.

It stands to reason that the products of a research deconstruction need to be tested by checking with readers, and by exploring with especially your study leader, the extent to which the set of deconstructed components as captured in your thesis or dissertation, is in line with the general usage and meaning of the components, while being articulated to the purpose and requirements of your research and contextualised to the scope and range of your research target group.

As is often the case with master’s and doctoral studies, the deconstructed information may apply more widely than just the target group for the study. The deconstructed data may not be limited to component meanings associated with only your abstracted categories as defined in your thesis or dissertation. How you group your data is up to you and you may test new concepts and their technical or academic definitions. The dominant logic of the process of deconstruction is abduction, although induction plays a part in testing the scope and range of the constructed concepts and their meaning in terms of a variety of related everyday meanings.

For the sake of efficiency, you will start with meaningful components that you already deconstructed previously. By linking subsets of components, according to plausible themes, which should be the problem statement or hypothesis of your research broken down into abstracted categories, you can produce a compact set of concepts and associated academic meanings articulated to the purpose of your research. These ‘sets of concepts’ are the typologies through which you communicate your arguments in a thesis or dissertation.

Typologies not only provide descriptions but also enable a clear exchange of deeper understanding about the meanings of words and concepts with which you work in your thesis or dissertation. Hence, typologies answer ‘what’ questions but not ‘why’ questions. Stated differently, your typologies reflect the ontology of your research, which you will need as the foundation for the epistemology, which would be your discussion, analysis and explanations of your arguments.

The epistemology of your thesis or dissertation proposes and tests discriminating insights about associations between elements of the regulatory and the primary ‘why’ questions. Because a theory or argument should at least hold across the same for your research, the testing should be applied to each unit of a selected sample to ensure validity with a reasonable probability of being accurate. You will not statistically calculate the probability that your sample is large enough to provide a good measure of accuracy when conducting qualitative research. However, you should take great pains in ensuring accuracy of your findings, for example by making your sample as large as possible, consulting as many different sources of information as you can reasonably obtain, asking readers for comment, arranging focus groups, etc.

Empirical generalisation

Empirical generalisation should not be confused with empiricism, which is a paradigm, as you should know by now. Empirical generalisation is studies based on the collection and presentation of evidence to prove a hypothesis or claim in the form of a problem statement or question. The evidence needs to be shown to be accurate, valid and credible. As such it represents the most basic requirements for qualitative research.

Empirical research mostly refers to evidence that can be observed and measured, which implies quantitative research. It can be directed at the ontology of a phenomenon, requiring you to focus on “what”, as well as the epistemology of phenomena, requiring answers to questions like “how many?”; “why?”; “what are the results?”; “what is the effect?”; and “what caused it?”.



  1. Is not an independent research method.
  2. Is used to group and articulate data to the purpose of research.
  3. Fits in well with constructivism.
  4. Can be rendered efficient through coding.
  5. Synthesises existing data to identify the inherent truth in the data.
  6. Needs to be checked by other stakeholders in the research.

On doctoral level, you will:

  1. Create new data from existing data.
  2. Escalate data to a higher level of abstraction.
  3. Develop or identify alternative meanings for words and concepts aligned with the problem statement, research question or hypothesis for your research.
  4. Mostly use induction.

On master’s degree level, you will:

  1. Deconstruct data to make it more creative.
  2. Mostly use deduction.

On doctoral and master’s degree level:

  1. Data need to be grouped into a set of categories and transformed into abstract types of philosophies and knowledge.
  2. You should aim at generalisation of your findings.
  3. You must ensure that your findings are logical, accurate and authentic.
  4. Typologies can be used:
    1. To communicate arguments in your thesis or dissertation.
    1. To describe concepts relevant to your research.
    1. To enable a clear exchange and deeper understanding of the meanings of concepts and words.
    1. To serve as an ontology upon which the epistemology for your research can be developed.

Empirical generalisation means providing solutions to a research problem, answers to a research question or evidence to prove or disprove a hypothesis.

Evidence must be accurate, valid and credible.


So, what do you think?

Is deconstruction just a euphemism for plagiarism?

Let’s put this question on ice for the time being.

The three articles following on this one deal with ethics.

Perhaps we will be in a better position to answer the question after we have taken a closer look at ethics and what it means.

Enjoy your studies.

Thank you.

Continue Reading