How to Conduct a Case Study

  • First of all, decide what it is you want to illustrate and what type of case study you might need to do so. Clearly, you might not know whether your case is deviant (or extreme) or representative when you begin but you should have a rough idea that by examining a particular person, phenomenon or place that you will shed light on the workings of your project or programme.
  • You then need to decide on the data collection methods. Often the case study approach involves the collection of multiple sources of evidence, using a range of quantitative methods (e.g., questionnaires, audits and analysis of routinely collected healthcare data) and qualitative techniques (e.g. interviews, focus groups and observations).
  • The aim is to provide as complete a picture as possible, perhaps using project documents (including meeting minutes), quarterly reports, midterm reviews, monitoring visits, mystery client reports, interviews, or observation. The use of multiple sources of data can increase the internal validity of a study.
  • You must hold on tight to the central or overarching question that you are trying to answer. That question will help you decide which information is relevant and which is not.
  • There may be a vast amount of data but only some of it is needed. You may be trying to show that your maths tuition project has improved the performance of secondary school pupils. For this, you may need a class (a control group) that has not had this tuition and one that has. You will need to consider the baseline results of both classes before the project, engage in interviews with pupils and tutors and identify exam results after one year of tuition.
  • The information which helps to show and explain any improvement (or lack of improvement) in performance will be central (e.g. interviews with pupils/ tutor groups whose marks went up and interviews with those whose marks went down). Can the findings from your target class be generalised more broadly to show that the tuition you are offering can make a real difference?

Ask Yourself

  1. Am I focusing on a person? If so, explain why you selected this particular individual to be studied and describe what experience he or she has had that provides an opportunity to advance new understandings about the research problem.
  2. Am I focusing on an incident or event? If so, does the event have a clear start and end date, and is it associated with an identifiable location? Is the event a rare or regular occurrence? Are you using it to test a hypothesis or illuminate new ways of thinking about an issue? To illustrate, you may focus on the outbreak of cholera in a particular village and use this to explore the effectiveness of your preparedness training and/or medical treatment programme in that village.
  3. Am I focusing on a place? Is that place unique or special in some way and, if so, can that uniqueness be used to build new understanding or knowledge about the question you are exploring? Why choose this place if similar places also exist? What are the attributes of this place that make it worth studying? E.g. a school catering for special needs in a remote part of South Sudan, a shelter for unwanted twins in Addis Ababa.
  4. Am I focusing on a phenomenon? To illustrate, the phenomenon could be the observation that roads have fallen into disrepair in parts of the Democratic Republic of Congo due in part to the fact that local farmers are sharpening their farm instruments on the tarmac surface. Your project may be about providing facilities and training in the maintenance of such equipment and/ or the maintenance of roads. You may need to undertake an in-depth analysis of a cause and effect that is grounded in an interactive relationship between people and their environment in some way.
  5. Am I focusing on more than one event, person, place or phenomenon? If so, are they ‘most similar’ or ‘most dissimilar.’

If they are most similar, you will expect limited variance and need to explain any that occurs. With most dissimilar cases, you may be more surprised by the similarities between the cases (e.g. Albinos in Benin and South Africa may face remarkably similar challenges despite the very different histories and racial records of these two countries)

NOTE: The choice of the case or set of cases to study cannot appear random. You may wish to refer to any prior studies that helped you determine that the case you chose was appropriate for investigating the research problem.

Problems to Avoid

Data overload:

Collecting large volumes of data that are not relevant to the case or too little to be of any value or too overwhelming.

Overgeneralisation:

Take care when drawing conclusions from your case study. They must be evidence-based and grounded in the results of the study; otherwise, they are merely speculation.

Failure to Document Limitation:

No case is going to reveal all that needs to be understood about a research problem. Therefore, just as you have to clearly state the limitations of a general research study, you must describe the specific limitations inherent in the subject of analysis. For example, the case of studying how men understand family planning in populous poor countries such as Mali may have limited application in other cultural contexts or in wealthier, less populated countries such as Namibia.

Failure to Extrapolate All Possible Implications:

Just as you should not overgeneralise from your case study conclusions, you also have to be thorough in the consideration of all possible outcomes or recommendations derived from your findings. If you do not, your reader may question the validity of your analysis, particularly if you leave gaps or fail to document an obvious causal factor or outcome from your case study research.

For example, if you are documenting the high incidence of fitsula problems in the Kivu regions of the Democratic Republic of Congo, you would have to include some analysis of the wider context: ongoing low-level conflict and the high incidence of rape by Congolese army soldiers and militiamen from many different rebel groups.