Data Analysis
In order to make the data that has been collected throughout the course of an M&E plan, presentable and utilised, a data analysis strategy should be implemented.
This is the final step in conducting monitoring and evaluation. In the written reports of the evaluation, there should be a specific section detailing how the data was reviewed. The analysis will enable the data that has been collected to become translatable in relation to the project’s logframe, its intended outcomes and objectives.
When analysing data collected it is important to ensure that the data is both valid and reliable.1 The validity of data can be understood to mean how accurate this information is. While reliability refers to how likely the same results will be produced when the same data collection method is repeated. These both should be considered when conducting data analysis.
The data analysis identified in a M&E plan will be dependent on the indicators and type of data that has been collected throughout the monitoring and evaluation process. Quantitative indicators and their data can be analysed and translated into graphs and statistics to reflect the outputs, outcomes and impact of a project.
Qualitative indicators and their analysis can prove to be a bit more challenging. This is because things such as answers to interview questions can be interpreted in different ways by different individuals.
It is important when conducting the data analysis to remain as objective as possible. If indicators have been selected based on considering the need for objectivity, it will translate to creating making objectivity in the analysis process more likely.
When analysing data such as informant interviews coding is a method that can assist in interpreting the results and relating it back to understanding the indicators. Coding can be utilised to organize the information and what came out from the conversation by identifying patterns. When conducting data analysis triangulation to validate the data and its results should be implemented by applying more than one method to review the data received.