Indicator matrices, the M&E blueprint

 

Tyryn Carnegie

In our first blog post, we looked at why a Theory of Change (ToC) is an important tool for road-mapping your organisation’s activities to its goals. The next blog post introduced indicators, measurements of how well your organisation’s actions are bringing about its goals. We also looked at SMART as a framework for deciding on what would be a good indicator. In this blog post we dive even deeper into indicators – we are discussing indicator matrices.

What is an indicator matrix?

An indicator matrix, like a ToC, is a tool. Its primary function is to systematically link your ToC objectives to indicators. This way, we can ensure that your organisation’s progress towards every outcome/goal is measured. An indicator matrix’s other function is mapping each indicator either to data that are already available or to data that need to be collected. In this way it highlights any data gaps that need to be filled.

Linking objectives to indicators

An indicator matrix is easiest presented on a spreadsheet. We continue with our example of a community health intervention:

 

Like with the ToC, we work backwards from your organisation’s goals. In this simple example, the goal is to see the community’s health outcomes improve by replacing wood and coal burning stoves with smoke-free stoves. To achieve its goal, the organisation has identified two objectives that must be met: firstly to replace all wood and coal burning stoves, and secondly to see a year on year decrease in lung diseases.

 

Now comes the crucial part: identifying relevant indicators that show us that these objectives are being met. As discussed in the previous blog post, these indicators should be selected based on the SMART criteria and this process usually requires a lot of brainstorming, desktop research, consultation with experts and sense-checking with each organisation to understand any constraints to data collection such as staffing and budget. Once each of the organisation’s objectives are linked to an indicator, we need to think about the data needed to measure these indicators.

Linking indicators to data sources

Let’s imagine that in our example the organisation, due to it being closely connected to the government health department, has data on the percentage of community members with each of the lung diseases.

 

From the above complete indicator matrix, we can quickly see that the organisation has two issues that need to be addressed.

 

The first is that there currently aren’t any household data being collected on the number of households using smoke-free stoves. The second is that the data needed for the indicator % community members with lung cancer are only available for every second year.

 

To address the first issue, the organisation will need to set up a household survey that asks households what type of stove they use. The second issue may be addressed by requesting that the government releases the lung cancer data for the missing years. If this is not possible, it may be necessary for the organisation to ask in their annual household survey if any household members are diagnosed with lung cancer. Otherwise, the average annual change in lung cancer could be calculated from the bi-annual data, but the organisation would only be able to report this indicator every two years.

Conclusion

An indicator matrix’s utility grows with the number of goals, outcomes and objectives that organisation has. Without one there is a good chance that an objective is left without an indicator, which is a big problem for M&E as it cannot be measured and thus cannot be evaluated. In our next few blog posts we’ll be looking at different methods of data collection in a broad but interesting and incredibly important topic: surveys.