There are many data collection methods available and deciding which to use can be daunting. Remember to choose data collection methods that will match your indicators. Participatory methods can be augmented by more formal data collection methods. These can complement or contribute to the information provided by your participants. Methods should be selected according to their usefulness in helping to understand impact. Table 11: Examples of matching data methods with outcome indicators Outcome indicator Data method (example) Data source (example) Enrolment rates for school-age children Document review (quantitative data) Statistics published by the Ministry of Education (secondary source) Proportion of women and girls in three communities who are aware of how to access support. Interviews, case studies or focus groups (qualitative data) Engaging with women and girls in the three communities (primary source) Proportion of men who agree that violence against women is never acceptable. Survey (quantitative data) Results of a survey of males over 18 in the population (primary source) Courts treatment of juvenile offenders Observation (qualitative data) Sitting in on court cases involving juvenile defenders (primary source) Your data collection method also needs to match the size, context and timeframe of your MEAL Plan. For example, a smaller activity should use fewer and more simple data collection methods than a longer activity. You want to gather enough data to be able to measure the impact of your NHRI’s work, while avoiding making the MEAL activity impractical, difficult or expensive. Consider aligning your data gathering with: • the purpose of the MEAL Plan, • the scope, length and complexity of the activity (the more complex it is, the more resources and time should be committed to the evaluation), • whether there are external stakeholders or funders who require accountability, • whether the MEAL activity would benefit from an external perspective and capacity, noting that i) external consultants can also being their own bias into an evaluation and ii) they would need to be familiar with the context of your NHRI. • whether there is a need to ‘triangulate’ information (validate the data by using more than one source or using more than one method). It is good practice to have the same data collection methods for establishing the baseline information (prior to the activity beginning), for monitoring (during the activity) and for evaluation (at the end of the activity). This makes it easier to compare results throughout the course of the activity. 63