Sampling error
Sampling error is the difference between a sample estimate and the population value caused by observing only part of a population.
What sampling error is
Sampling error is the difference between an estimate from a sample and the true population value that the sample is trying to estimate. It exists because the study observes only some members of the population, not every member.
Why it happens
Even a well-designed random sample will not perfectly match the population. One sample may include slightly more younger people, high earners, or frequent users than another sample. Those random differences can move the sample statistic away from the population parameter.
Sampling error versus bias
Sampling error is not the same as bias. Sampling error is the random variation that comes from using a sample. Bias is a systematic problem, such as undercoverage, nonresponse, leading questions, measurement error, or a sampling frame that leaves out part of the target population.
How it is measured
Sampling error is usually summarized through standard errors, confidence intervals, or margins of error. These tools do not reveal the exact error in one estimate, because the true population value is usually unknown. They describe how much estimates would be expected to vary under the sampling method.
Sample size
Larger samples generally reduce sampling error because they give more information about the population. The improvement is not one-for-one: cutting a standard error in half often requires about four times as many independent observations, assuming the design and variability stay similar.
Survey design
Real surveys often use stratification, clustering, weights, and complex selection rules. These design choices can improve coverage or operational efficiency, but they also affect sampling error. Official statistics agencies often publish methods, standard errors, or margins of error so users can account for the design.
Nonsampling error
Nonsampling error covers problems other than random sampling variation, including nonresponse, coverage gaps, interviewer effects, processing mistakes, and inaccurate answers. Increasing sample size can reduce sampling error, but it does not automatically reduce nonsampling error.
Why it matters
Sampling error reminds readers that a sample estimate is not an exact population fact. It helps prevent overreading small differences in polls, public statistics, market research, health surveys, and any analysis that generalizes from a sample to a larger group.