D annan a simple guide to resarch writing indd

34 
 

---

 
 
 
Data collection methods in the case study
 
 
Figure 3.3: Case study data collection methods
 
The case study design differs vastly from other research methodologies in that researcher 
tries to examine a huge number of elements by constraining the number of cases that 
should be observed. Another distinctive element is that the case study researcher in 
general tends to be an in-depth investigator of phenomena, as they exist in the character 
setting (Yin 2009).
Two techniques that are often used to analyse case study data are pattern matching and 
time-series analysis. Pattern matching is the process of taking several pieces of information 
that seem logically related and connecting them to some idea or theoretical proposition. 
The patterns should be compared to rival theories, looking for the best ‘fit’ of observed data 
to theory. Time-series analysis involves comparing a trend of data points to some 
theoretically significant trend, or to some other alternative explanation.
Action research
 
Action research can be defined as a reflective process of progressive problem solving 
facilitated by individuals, or groups working with other teams trying to improve and address 
an issue in a community. Action research is a strategy that has the dual aims of action and 
research:

35 
 

---

 
 
 
• The action is to bring about change in for instance a community or department
• The research seeks to increase the researcher’s understanding and/or that of the 
target group/community.
Action research is a cyclical process that consists of:
• planning
• action
• review of the outcomes of action
• reflection because of outcomes.
At each point in this cycle, the data available is used to determine the next stage. The 
whole purpose of the research is to determine simultaneously an understanding of the 
social system and the best change opportunity available.
Researchers generally adopt an action research design to increase the reliability of their 
findings by working with various data sources, ideally autonomous or partly independent 
samples of witness/informants, diverse research settings, the same informant responding 
to different questions which address the same topic from some different directions, data 
gathered at various occasions by different researchers or diverse techniques.