If external validity is threatened by a study's sample being unrepresentative, which practice would help protect it?

• A. Increase the sample size without changing sampling method
• B. Use probability-based random sampling
• C. Restrict the study to a single cohort
• D. Use a Convenience Sample

Answer: (B)

Protecting external validity means making sure the sample reflects the population you want to generalize to. Using probability-based random sampling achieves that by giving every member of the population a known, nonzero chance of being selected. This approach reduces selection bias and, especially when you use techniques like stratification or cluster sampling, helps ensure subgroups are represented in the sample in proportion to the population. With a properly selected random sample, the results are more likely to generalize beyond the study participants.

Increasing the sample size without changing how participants are chosen doesn’t fix a biased selection process—the sample can still be unrepresentative even if it’s large. Restricting the study to a single cohort narrows the scope and undermines generalizability. Convenience sampling, while easy, tends to produce biased samples that do not reflect the broader population.