Looking briefly at their rating methodology guide, I am cautiously impressed with their approach to impact evaluation, considering that they try to perform shallow evaluations at a large scale, at least within-cause. Specifically, it seems to me that they understand the limitations of their models well.
That said, between-cause comparison doesn’t exist, the quality of evidence isn’t taken into account as far as I can tell, and there probably are some other issues.
Looking briefly at their rating methodology guide, I am cautiously impressed with their approach to impact evaluation, considering that they try to perform shallow evaluations at a large scale, at least within-cause. Specifically, it seems to me that they understand the limitations of their models well.
That said, between-cause comparison doesn’t exist, the quality of evidence isn’t taken into account as far as I can tell, and there probably are some other issues.
I’ll try and take a closer look 🕵️♂️