Presentation is loading. Please wait.

Presentation is loading. Please wait.

Pieter Blignaut according to Hofstee (2006)

Similar presentations


Presentation on theme: "Pieter Blignaut according to Hofstee (2006)"— Presentation transcript:

1 Pieter Blignaut according to Hofstee (2006)
The Method Pieter Blignaut according to Hofstee (2006) 15 March 2014 Hofstee, E. (2006). Constructing a Good Dissertation. EPE: Sandton

2 Agenda Role of the Method Chapter Layout Designing a Method
Common Research Designs

3 Role of the Method Problem statement Method Results Conclusions
A result can only be accepted / rejected / checked / replicated / understood in the context of how you got there. Choose a design to be appropriate to your work. What data? How collected? How analysed? Isolate the phenomenon that you are studying Control for external factors that could affect your results Ronnie

4 Ethical considerations
Chapter Layout Introduction Research design Methodology Instruments Data Analysis Limitations Ronnie Ethical considerations Conclusion Hofstee, E. (2006). Constructing a Good Dissertation. EPE: Sandton

5 Introduction Section All chapters need an introduction
Reiterate the purpose of the study Overview of what will follow in the chapter How does the content in this chapter link with the content in the previous chapter. What do you want to achieve with your method? Make it clear – not a meal. Match with the content in Chapter 1, but don’t repeat verbatim. Give the reader an overview. Ronnie

6 Research Design Overall approach to data capturing and analysis
Provides theoretical background to your methodology An established research design provides readers with a conceptual framework. Get detailed information on a research design before you commit yourself Chosen research design(s) must be appropriate for your problem statement No details or specific content of your techniques – must go to Methodology chapter You may use more than one design if they will complement one another, e.g. surveys and experiments Don’t go overboard Content Name the research design(s) you will use Overview of the theory Motivate the choice(s) Also motivate why you did not use some others Discuss the strengths and weaknesses as it relates to your study Ronnie

7 Common Research Designs
Extended Literature Reviews Surveys Case Studies Correlation-Based Research Comparisons Content Analysis Critical Theories Historical Studies Evaluative Research Ethnographic Research Action Research Experiments Secondary Data Analysis Simulations Interdisciplinary Research Theory Development Ronnie

8 Research Design Example problem statements for which we may need a design: How does font colour affect reading speed? Does soft drinks shorten the life span of those who drink them? Does playing baroque music to cows during milking raise the daily yield? Does poverty in Africa has a worst effect on life expectation than poverty in Europe? Ronnie

9 Common Research Designs
1. Extended Literature Reviews Provide an overview of a field Takes a lot of reading Can provide new perspectives on existing knowledge Show links between subfields 2. Surveys Elicit information from individuals who have the knowledge you are seeking are able and willing to cooperate are representative of a larger group Find out people’s opinions, desires and attitudes Potential minefield May lead to questionable research Ronnie

10 Common Research Designs
3. Case Studies Examine something that happened or an organisation in a structured way Useful when detailed information is required of any particular case Difficult to obtain unbiased results Observational studies Draw interference about the effect of a treatment or factor on dependent variables Be aware of ethical issues Control for external factors Ronnie

11 Common Research Designs
4. Correlation-Based Research Compares two or more variables to establish if there is a relationship between them, e.g. smoke and heart attacks Relationship can be positive or negative Be careful of outside factors Difficult to obtain reliable and representative data Correlation is not necessarily causation Longitudinal studies Repeated observations of the same variables over time Allows distinctions between short-term and long-term phenomena Cohort studies select a group that shares an event, e.g. birth Panel studies samples a representative cross-section Retrospective studies look back in time Cross-sectional studies Observations of different individuals from a representative sample with respect to relevant factors while controlling for external factors Ronnie

12 Common Research Designs
5. Comparative Analysis Compare two items in a focused and systematic manner The broader the comparison, the more difficult is the study Suitable in an experimental setting where most variables are under control of the researcher Difficult in the social sciences and humanities where it is difficult to control variables 6. Content Analysis Examine the content of documents, films and videos Goal is to discover the non-obvious meaning contained in the document Mostly quantitative Applicable when an in-depth understanding of text is required. Ronnie

13 Common Research Designs
7. Critical Theories Takes a critical perspective on society Explicitly political Historical Studies Looks at an event in the past to shed light on the event or on a contemporary issue Likely to be a tool to strengthen other parts of your work Evaluative research Evaluates the effect or success level of some intervention Can be quantitative or qualitative E.g. cost/benefit analysis Be careful to establish appropriate measurement criteria Need sufficient evidence to be convincing Acknowledge the limitations of interpretation Ronnie

14 Common Research Designs
10. Ethnographic Research Observe participants in their natural setting to understand behaviour Danger of subjectivity Difficult to write up results 11. Action Research Participants are actively involved Learning by doing Individuals work with others in a team to solve problems Emphasis on planning, implementation, learning and empowerment of participants Applicable when there is a small group of participants in a working environment who need to achieve a common set of goals Ronnie

15 Common Research Designs
12. Experiments Test a hypothesis or theory or observe the effect of a given intervention Record and explain outcomes This is the classic design because of replicability Can be done in lab or field Lab experiments allow control of the environment and can easier be replicated Cause and consequence “If you do the same thing in the same environment you will get the same results” Control for outside variables Measurements must be accurate Not always possible to set up reliable experiments for real-world scenarios Easier to prove an expectation than to explain why it happened Don’t try to be innovative – stick with the traditional way of doing things. Ronnie

16 Common Research Designs
13. Secondary Data Analysis Study data that was collected previously to check answer questions Can be quantitative or qualitative Huge amount of data available, e.g. collected by government agencies, financial data, weather data, web sites, etc. Important to know that the data is reliable 14. Simulations / Modelling Capture essence and key variables of a process and create representation Statistical model describes a process Simulation runs similar scenarios, mostly on computer A successful model can explain much about a particular phenomenon Large practical an theoretical implications Limited: No model is reality Can be complicated Ronnie

17 Common Research Designs
15. Interdisciplinary Research Takes methods, concepts, ideas from one discipline and applies them to a problem in another discipline Difficult: Expects mastery of two disciplines Creativity and common sense needed Can have theoretical and practical significance Important to limit the scope and build on previous work 16. Theory Development Test an aspect of existing theory Use various models and data to achieve, e.g. mathematical modelling and philosophical reasoning Rewarding to develop a new theory that contributes to human understanding of nature Ronnie

18 Select a Design Brainstorm Get a list of possible designs
Get ideas from others Ask: Is it appropriate for your problem statement? Are you familiar with the design? Will it be possible to obtain data? Will the data be reliable? Will you be able to analyse the data? Will you be able to keep to your time frame? Ronnie

19 Ethical considerations
Chapter Layout Introduction Research design Methodology Instruments Data Analysis Limitations Ronnie Ethical considerations Conclusion Hofstee, E. (2006). Constructing a Good Dissertation. EPE: Sandton

20 Methodology Details of the research design
How are you going to implement the selected design? Instruments Anything that you will use to capture data E.g. Questionnaires, Lab equipment, Software tools Self development or use of existing materials? Purpose of the instruments. What were they originally intended for and what do you hope to achieve with them? Reliability and validity of the instruments Ronnie If your instruments are questionable, so is your data. If your data is questionable, so are your results. If your results are questionable, so are your conclusions. If your conclusions are questionable, so is your PhD / M.

21 Methodology Data Discuss the strengths and weaknesses of the data that will be captured with your method Convince the reader that the data will be of sufficient quality to allow analysis and obtain valid results. What are the possibilities that your data will be flawed, incomplete or biased? Describe the characteristics of the population and sample. How representative will your data be? How did you arrive at the sample? Analysis You need to turn data into evidence. How? Statistical techniques to be used. You must make a case that what you do will lead to the knowledge that you are after. Ronnie

22 Ethical considerations
Chapter Layout Introduction Research design Methodology Instruments Data Analysis Limitations Ronnie Ethical considerations Conclusion Hofstee, E. (2006). Constructing a Good Dissertation. EPE: Sandton

23 Limitations and Challenges
All methods have limitations Name the limitations that will affect the reliability of your findings limit the extent to which you can generalise What are the consequences of the limitations and why will your findings still be worthwhile? Challenges E.g. only 15% of respondents completed answered a specific question in the survey form. How are you going to deal with it? Limited access to critical documentation / information. Ronnie

24 Ethical Considerations
Inform the reader about the ethical issues that had to be considered, e.g. health and safety of participants anonymity of participants sensitive information environmental issues, e.g. disposal of electronic equipment Did you get ethical clearance from the university? How did you address the challenges? Ronnie

25 Ethical considerations
Chapter Layout Introduction Research design Methodology Instruments Data Analysis Limitations Ronnie Ethical considerations Conclusion Hofstee, E. (2006). Constructing a Good Dissertation. EPE: Sandton

26 Conclusion Section Overview of what was covered in the chapter
Chosen research design What is next? This chapter should lead into data capture, results and analysis. Only one or two paragraphs to round off the chapter. Ronnie

27 Concluding Remarks Put yourself in the shoes of the reader
Don’t write from the perspective of somebody who knows what’s going on (you). Make sure that you covered all details. This is a thesis/dissertation – you have enough space. Return to the chapter if you made changes to the design along the way. PILOT!! Ronnie

28 Thank you


Download ppt "Pieter Blignaut according to Hofstee (2006)"

Similar presentations


Ads by Google