Explain sampling bias with an example.

Explain sampling bias with an example.

• A. Bias arises when the sample perfectly represents the population.
• B. Bias arises due to overrepresenting the population by using the entire population.
• C. Bias is introduced when the sample is not representative of the population.
• D. Bias can be avoided by increasing sample size only.

Answer: (C)

Sampling bias happens when the people you study don’t reflect the broader group you want to understand. The error isn’t just about how many responses you collect, but about who ends up in your sample. Consider estimating how many city residents support a new bicycle lane. If you only survey people who ride bikes to work, you’ll likely get higher support than the overall population, because cyclists aren’t representative of all residents. This happens because the method of choosing participants favors a particular subgroup, so the results won’t generalize to everyone living in the city. That’s why the concept matters: it’s a systematic mismatch between the sample and the population. If the sample truly represented the whole population, there would be no bias. If you could survey the entire population, sampling bias would vanish. Merely increasing the number of responses doesn’t fix bias, because the problem lies in who is included, not how many are included. To avoid bias, you need a sampling method that covers all groups fairly—random sampling, stratified sampling, or ensuring the sampling frame includes all relevant subgroups.

Sampling bias happens when the people you study don’t reflect the broader group you want to understand. The error isn’t just about how many responses you collect, but about who ends up in your sample.

Consider estimating how many city residents support a new bicycle lane. If you only survey people who ride bikes to work, you’ll likely get higher support than the overall population, because cyclists aren’t representative of all residents. This happens because the method of choosing participants favors a particular subgroup, so the results won’t generalize to everyone living in the city.

That’s why the concept matters: it’s a systematic mismatch between the sample and the population. If the sample truly represented the whole population, there would be no bias. If you could survey the entire population, sampling bias would vanish. Merely increasing the number of responses doesn’t fix bias, because the problem lies in who is included, not how many are included. To avoid bias, you need a sampling method that covers all groups fairly—random sampling, stratified sampling, or ensuring the sampling frame includes all relevant subgroups.