We've just completed version 1.0 of EvalC3: a tool to help investigate causes of success and failure in development aid projects.

Using EvalC3 you can analyse monitoring and evaluation data that describe how projects where carried out (what activities were performed, or what other conditions were met) and whether their goals were achieved. In particular it helps automate the search for combinations of conditions that correlate well with achievement of desired outcomes. That's just the sort of analysis you want if you are managing a portfolio of development projects and you want to understand what works.

And EvalC3 isn't limited to M&E for development projects, you can use it to investigate any data where you want to understand what combinations of conditions correlate with achievement of results.

EvalC3 is an Excel application that helps you analyse your data. Paste in your own data and tell it which columns are attributes or conditions, and which corresponds to achievement of the outcome. Behind the scenes it creates all the necessary formulas to evaluate different combinations (or configurations) of attributes as predictors of the outcome using a variety of standard performance measures. You can configure attributes manually and the performance measures update immediately. And EvalC3 also automates searching for the best performing configurations.

EvalC3 is being developed by monitoring and evaluation expert Rick Davies; he describes the theory and motivation for it in this presentation. It will be available soon via evalc3.net.