Which statement is true about sampling irrespective of sample size?

• A. The sample bias is reduced if the sample being selected is representative of the population.
• B. Larger samples always eliminate bias.
• C. A random sample guarantees no bias.
• D. Sampling error is independent of sample size.

Answer: (A)

The main idea here is how bias arises from the way a sample reflects the population. If the sample is representative of the population, the bias from the selection process is minimized. In other words, when the sample mirrors the population on the key characteristics that matter for the estimate, the chances that the sample estimate will systematically differ from the true population value are reduced. This holds true regardless of how many observations you have, because representativeness directly addresses the source of sampling bias.

Keep in mind that even a representative sample isn’t immune to all errors. Other issues like nonresponse, measurement mistakes, or respondent bias can still creep in, but the core risk of selection bias is lowered when the sample matches the population.

The other statements miss this distinction. Larger samples reduce sampling error and improve precision, but they don’t automatically remove bias caused by flawed selection or design. A random sample lowers the risk of selection bias, but it doesn’t guarantee no bias at all—bias can still occur from nonresponse or measurement issues. And sampling error is closely tied to sample size: as the sample grows, sampling error typically decreases, not remains independent of size.