- A kind of bias that arises in taking samples, especially in cluster randomized trials. An additional observation drawn from the same cluster is likely to be more similar to others in the cluster than if drawn randomly from the whole population. The design effect indicates how much larger a sample is needed to achieve the same precision as a simple random sample. It is the ratio of the true variance of a statistic (taking the sampling design into account) to the variance of the statistic for a simple random sample with the same number of cases.
- acronymDEFF (written as design effect)