Measures and Metrics
The measures chosen should be set to SMART criteria.
When we are measuring impact against our theory for change, we use different measures for each stage. That is: we can use impact measures, outcome measures and output measures. Outputs influence outcomes, and outcomes in turn, influence the impact we want to see. Together, these may bring us closer to our greater vision for change.
Here, we use the terms ‘measures’ and ‘indicators’ interchangeably.
Impact measures are indicators that are linked to the objectives of your intervention, and are generally achieved in the medium-to-long term. Some examples of impact indicators in the area of infant health are¹:
- % of under-weight under three years (or under five years)
- % of low birth weight infants
- Incidence of diseases that have an impact on nutrition (Malaria, diarrhea, acute respiratory infection, and HIV/AIDS)
Outcome measures are indicators that are linked to results. Some examples of outcome indicators in the area of infant health are²:
- % of health providers with knowledge of key nutrition messages and actions at critical stages in the life cycle of women and children
- % of health facilities without any stock of Folic Acid or Vitamin A
- Type and coverage of community-based initiatives to promote food diversification including food production, processing, preparation and preservation of available resources
Output measures are indicators linked to the activities of your intervention (process, product or service). They monitor the immediate outputs of an intervention (process, product or service). Some examples of output indicators in the area of infant health are³:
- % of infants that were breastfed within one hour after delivery
- Use of iodized salt
- % of postnatal women receiving Vitamin A supplements (200,000 IU) within 8 weeks of delivery
You may notice that each of these indicators for impact, outcome and outputs are a combination of quantitative and qualitative data. It’s always important to collect a mixture of both in your measurements.
Measuring for system change needs to focus on 2 things:
- Measuring change
- Measuring to change
Generally, indicators can either be lagging or leading. Lagging Indicators are those that follow an event. They measure change (tell you what is happening or not happening), but don’t tell you what to change. An example is weight. Leading indicators are those that signal future events such as calorie intake. They are measures to change, generally more likely to tell you what you ought change in order to improve your results. Leading indicators are often not very helpful about telling you about what to change. Depending on the context of your challenge, you will (and should) have a mixture of both leading and lagging indicators.
Often process or input indicators give you better clues about your interventions but are not usually used in a large-scale systems. Outcomes or lagging indicators tell you about the state of affairs, which is useful to indicate if there’s been an effect. Ultimately, you need to measure change, and to do that, you need to measure outcomes. In order to get better at bringing about change – a combination of both leading and lagging indicators gives you a better sense of if you’re realizing change, and what you need to change in the interventions you're working on, because true evaluation is also about: How did you get here?
A metric is a composite of indicators and is a number that tells you important information about a state.
Indicators chosen should be set to SMART criteria:
- Specific: Clear and well-defined target for improvement
- Measurable: Quality or indicator that we can somehow gain value from
- Achievable: Outcome should be feasible
- Realistic: In terms of what can be done in terms of evaluation (e.g. feasibility due to resources etc.)
- Timely: Specify when you expect to see results.