As a think tank communicator, I often wonder what are the best ways to share evidence-based research with a diverse audience. At the Public Affairs Centre (PAC) we have realised usefulness of including a data analytics component in all the research that we undertake. For instance, in a study to understand why stunting in children occurs (despite the government’s considerable effort to address this problem) interesting findings emerged: Through the anganwadis (rural child care centres in India- the literal translation is ‘courtyard shelter’) the government provides iron tablets to pregnant women. Despite this, many children are underweight at birth. PAC conducted a brief research study, and the data analysis revealed that women were not actually taking the iron tablets – they would collect them but not use them- due to superstitious beliefs arond the side effects. This insight guided the IEC campaign strategy that followed.
Data analysis is the backbone of any research. Researchers often collect a huge amount of data. However, the process of cleaning, transforming, and modelling the data with the objective of drawing useful insights is the key- otherwise the data is useless.
PAC’s experience with data analysis
- Splits macro problems into micro units
- Structures findings from different sources
- Sifts through meaningful insights
- Eliminates subjectivity
Data analysis breaks down complex information into a simple definition and provides evidence and an increased understanding of conducted research. Ideally, a good data analysis ensures a process to scrutinise, sift change and perhaps transform and amend information to arrive at a realistic conclusion.
Importance of analysing data
In recent times many research organisations have realized the utility and importance of including data analytics in research programmes because it:
- Provides credibility to research
- Interprets information in a very clear manner to ensure that a diverse audience understands information clearly
- Backs up data and provides a theoretical base
- Provides a clear insight and interpretation
What is data interpretation?
Data interpretation typically means adhering to a process where data is examined and analysed to ensure a logical interpretation. The most common interpretation is the combination of qualitative and quantitative analysis.
While quantitative data tells you what is happening, the rate at which it is happening and the patterns of incidence, it is qualitative data, if garnered and used intelligently, that explains why it is happening. To know what the data is trying to tell you, it helps to draw on qualitative data. Qualitative data is the analysis of non-numerical data, like text or oral communication, and includes greater depth and breadth of information. On the other hand, quantitative analysis focuses on the measurement using statistics to identify results.
To keep in mind
In order to ensure better informed decisions and sound analysis of data, the following must be kept in mind:
- Gather data to reflect completeness and genuiness
- Select qualitative or quantitative analysis or both
- Reflect and evaluate if the data meets the stakeholder’s requirement
- Identify gaps – inaccuracy in data, bias arising from being subjective, correlation versus causation.