Sometimes referred to as experimental evaluations, randomized controlled trials or RCTs randomly allocate potential beneficiaries of an intervention to a program or treatment group (who receive the intervention) or a control group (who do not). Outcomes for the two groups are then compared.
They are most often used to test medicines or medical procedures, but they are becoming more common in social interventions, particularly in relation to early years programs and education interventions in the US.
RCTs are considered the most reliable way of testing the effect of an intervention on outcomes for the potential beneficiary. Since the subjects of a trial are allocated at random to program and control groups, both are statistically equivalent and comparisons of outcome will reflect the effect of the intervention and not the characteristic of the groups.
Most importantly, RCTs eliminate selection effects. For example, if entry to the program tested was not random, the outcome might be the result of one group wanting the intervention more than another.
RCTs are strong at estimating the size of the difference in predefined outcomes between program and control groups. It is possible, therefore, to estimate how much change is the result of the intervention.
Other evaluation designs, including quasi-experimental designs that include a control group can detect associations between an intervention and an outcome but they cannot rule out the possibility that the association was caused by a third factor linked to both.