Which statement best defines clustered data?

• A. Observations that share common features not taken into account by covariates in the model
• B. Observations that are independently sampled without shared features
• C. Data where all units have identical covariate values
• D. Data collected from a single measurement per unit

Answer: (A)

Clustering happens when observations share common features within groups, so those within the same group are more similar to each other than to observations from other groups. This creates dependence among observations that isn’t captured by the measured covariates, because unmeasured group-level factors influence all members of that cluster.

The statement that best defines clustered data says exactly that: observations share common features not taken into account by covariates in the model. In practice, this means members of the same cluster (like patients within the same hospital or students within the same school) tend to be more alike due to shared environment, practices, or other unmeasured characteristics, leading to correlated outcomes.

The other ideas don’t capture clustering: independent sampling with no shared features would not exhibit within-cluster correlation; having identical covariate values across units removes between-unit variation but doesn’t describe groups with shared, unmeasured influences; data from a single measurement per unit doesn’t address within-group dependence.