samplingfout

Samplingfout is the Dutch term for sampling error, the discrepancy between a statistic calculated from a sample and the corresponding population parameter caused by the fact that the sample represents only a subset of the population. It arises even when measurements are precise and is distinct from measurement error or data entry mistakes. For an estimator θ̂ of a parameter θ, the samplingfout is θ̂ − θ. The sampling distribution describes how θ̂ would vary across repeated samples, and the standard deviation of this distribution is called the standard error of θ̂.

A common example is the sample mean x̄ estimating the population mean μ. Although E[x̄] = μ, a particular sample yields an x̄ that may differ from μ. If the population standard deviation σ is known, the standard error of x̄ is σ/√n; if not, it is s/√n, where n is the sample size and s is the sample standard deviation. By the central limit theorem, x̄ is approximately normally distributed for large samples, enabling confidence intervals such as x̄ ± zα/2 (σ/√n) or x̄ ± tα/2 (s/√n).

In practice, samplingfout is reflected in margins of error in polls and surveys. For example, a random sample of 1,000 voters might yield a margin of error of about ±3 percentage points for a proportion at 95% confidence. Samplingfout can be reduced by increasing sample size, employing proper random or probability sampling, using stratification, and applying weighting to adjust for sampling frame imperfections and nonresponse. The concept is closely related to, but distinct from, samplingbias (systematic errors due to nonrandom sampling) and other design effects.