STAGE 6. Defining the evaluation plan

What steps should be completed?

Step 1: Select the type of evaluation

RANDOM ALLOCATION

Random allocation defines the procedure by which individuals (or other sample units, such as schools or neighbourhoods) who are the subjects of an intervention are randomly assigned to the experimental condition (intervention group) or the control group to ensure that both groups are equivalent, and therefore comparable to each other.

Applying this procedure implies that both groups are similar, at least in relation to the personal and social characteristics relevant to the intervention so, if differences are found in the measurements after the intervention, these differences can be attributed to the intervention.

If the intervention group and the control group are not equivalent, selection bias arises. This is where differences between the groups could influence the results of the intervention, so it is possible that changes allegedly derived from the intervention are explained, at least in part, by existing differences between the two groups.

This procedure (random allocation to the experimental or control condition) and the fact that the intervention has a control group are the basic criteria to consider that an outcome evaluation has a robust design that ensures (infers with a high probability) that the results obtained are due to the intervention.

However, although random allocation is a procedure that implies good methodological rigour to demonstrate that an intervention allows the expected effect to be achieved, it can cause ethical conflicts that must be considered: if we develop an intervention in a target population it is because we believe that it will have beneficial effects on the individuals who are part of it, either because we are relying on previous scientific evidence, or because we are trying to confirm the hypothesis that this intervention yields the aforementioned positive effects. When assigning subjects to the control group, we are excluding them from these benefits. This ethical issue has to be addressed, explained and explained adequately in the evaluation plan.