Given Hapunda

162 individuals in order to achieve some sort of reliability. Having other people to review what has been analysed is important because it enhances and helps findings to be considered more useful and trustworthy. It is usually difficult to take as truth, work that has not been reviewed or seen from one eyes’ perspective. Whatever method of analysis and interpretation thereof, the aim of every evaluation is to produce good quality findings. Importance of Qualitative Data Analysis Some of the identified significance of conducting qualitative date analyses includes; a) Identifying any significant change that may have occurred in people and communities. b) Understanding any subtle indicators of social change that may have emerged from the data. c) Identifying ways that could improve the implemented programme. d) Gaining knowledge about emerging issues that may help understand the data. e) Enriching findings with lively and detailed information that lacks in quantitative data. f) Understanding culture, experiences and activities of diverse community members and the context of people’s lives. g) Understanding the community dialogue of who is included and excluded. Types of Qualitative Data Analyses Many forms of conducting analyses qualitatively exist (i.e. case study approach, theory-based approaches, and collaborative and participatory analysis). Therefore, some of the ways of organising, managing and analysing qualitative data are discussed. The Process Qualitative data from M&E methods like most significant change stories and focus group discussions are often messy and unstructured. Unlike quantitative data analysis, qualitative data analysis does not happen in a linear way, neither is it a neat or simple process. It requires a repeated process of critically reading, interpreting and reaching shared understanding of the data. To ensure effectiveness of data usage, it is imperative to generate a data organisation, management and analysis system that all those involved in the analysis follow. Coding Coding involves classifying or categorising individual pieces of data. Boyatzis (1998) states that a good code has five elements: