"In a Randomised Control Trial (RCT) different units are randomly assigned to separate groups. One group receives a development intervention and the other does not. Changes in the two groups over time are then compared to accurately measure the effect of the intervention. RCTs have been much debated ... over the past 10 years. Some see them as the ‘gold standard’ for impact assessment." (Introduction) more

"Sampling is a process that enables information to be collected from a small number of individuals or organisations within a project or programme, and then used to draw conclusions about a wider population. There are many different sampling methods. Quantitative analysis tends to require large, rand ... om samples. Qualitative analysis usually relies more on smaller, purposefully chosen samples." (Introduction) more

"Experimental approaches work by comparing changes in a group that receives a development intervention with a group that does not. The difference is then attributed to the intervention. In a full experimental approach, units are randomly allocated to two groups – one that receives the intervention ... and one that does not. In a quasi-experimental approach non-random methods of assignment are used instead." (Introduction) more