Which practice mitigates selection bias?

Which practice mitigates selection bias?

• A. Collect data from the most convenient sources.
• B. Rely on data from a single department only.
• C. Exclude any data that doesn't fit the hypothesis.
• D. Use representative sampling methods to ensure sample reflects population.

Answer: (D)

Mitigating selection bias comes from making sure the data you analyze truly represents the population you’re studying. Using representative sampling means applying methods like random or stratified sampling so the sample mirrors key characteristics of the population (such as job levels, locations, or demographics). When the sample reflects the whole group, the conclusions are more likely to be accurate and generalizable, not skewed by who happened to be chosen. The other approaches lead to bias. Collecting data from the most convenient sources tends to pick a nonrandom subset that may differ in important ways from the broader population. Relying on data from a single department ignores variation across the organization and limits generalizability. Excluding data that doesn’t fit the hypothesis amounts to cherry-picking and undermines the validity of the analysis. Using representative sampling directly addresses these issues, making it the best practice to reduce selection bias.

Mitigating selection bias comes from making sure the data you analyze truly represents the population you’re studying. Using representative sampling means applying methods like random or stratified sampling so the sample mirrors key characteristics of the population (such as job levels, locations, or demographics). When the sample reflects the whole group, the conclusions are more likely to be accurate and generalizable, not skewed by who happened to be chosen.

The other approaches lead to bias. Collecting data from the most convenient sources tends to pick a nonrandom subset that may differ in important ways from the broader population. Relying on data from a single department ignores variation across the organization and limits generalizability. Excluding data that doesn’t fit the hypothesis amounts to cherry-picking and undermines the validity of the analysis. Using representative sampling directly addresses these issues, making it the best practice to reduce selection bias.