A randomized controlled trial (RCT) is a type of quantitative scientific study widely used in medicine and development to gain evidence about the impacts of interventions.
In an RCT, participants are randomly split into a treatment group (who receive the intervention) and a control group (who do not). Afterwards, the traits of the two groups are compared, and any difference that did not exist beforehand is taken as evidence of the intervention’s effect. For instance, to learn about the impact of mosquito nets, researchers might randomly distribute the nets to some portion of the households in a village, and then afterwards compare the rates of malaria in the households that received nets and the households that did not.
Randomized controlled trials are often considered the “gold standard” of quantitative studies, since they avoid a number of problems such as “selection bias” and “confounding” which may arise in other types of studies. However, RCTs do have a number of limitations: they are usually expensive, their results may not generalize to groups unlike the ones tested, and withholding a treatment from the control group can sometimes be unethical.
Karnofsky, Holden. 2016. How we evaluate a study.
How GiveWell decide how much weight to place on an academic study.
Wikipedia. 2016. Randomized controlled trial.