Course contentsShow
← Data Science
Lesson 248 of 2,1457. Sampling and EstimationFree lesson

Sampling Error vs Sampling Bias

Distinguishing between random variation inherent in sampling and systematic errors that don't decrease with sample size.

Sampling Error vs Sampling Bias

You'll learn: How to distinguish between the unavoidable random variation in samples and the systematic problems that wreck your conclusions no matter how large your sample gets.

The Core Difference

Imagine you're trying to estimate the average height of all adults in your city by measuring samples.

Sampling error is the natural, random variation you get just because you didn't measure everyone. If you randomly sample 100 people today and another random 100 tomorrow, you'll get slightly different averages—that's sampling error. It's not a mistake; it's an inherent feature of working with samples. The good news? Sampling error decreases as your sample size increases. With 10,000 people instead of 100, your estimate gets much closer to the true population value.

Sampling bias is a systematic error in how you collect your sample that pushes your results in one direction, away from the truth. If you measure heights only at a basketball court, you'll consistently overestimate average height. If you only survey people during work hours (missing shift workers), you'll systematically miss part of the population. No amount of additional sample size fixes bias—you could survey 100,000 basketball players and still badly overestimate the population's average height.

The Key Distinction

  • Sampling error = random, decreases with larger samples, unavoidable but manageable
  • Sampling bias = systematic, persists regardless of sample size, comes from flawed sampling methods

Think of sampling error as the "fuzziness" around your estimate—it creates uncertainty but doesn't mislead you in a particular direction. Sampling bias is like having a broken scale that always adds 5 pounds—more measurements won't help; you need to fix the scale.

Key Takeaway: Sampling error is random variation that shrinks with larger samples; sampling bias is a systematic flaw in your sampling method that remains no matter how much data you collect.