1. The document discusses selection bias and randomized controlled trials (RCTs) for evaluating intervention effects. 2. Selection bias occurs when the groups being compared have inherent potential differences, making it difficult to accurately estimate the true intervention effect. An RCT helps reduce this issue by randomly assigning subjects to intervention and control groups. 3. In practice, an RCT may not always be feasible due to costs or other limitations. The document explores alternative methods for estimating intervention effects when selection bias is present.