Course contentsShow
← Data Science
Lesson 258 of 2,1457. Sampling and EstimationFree lesson

Comparing Population, Sample, and Sampling Distributions

Distinguishing between the three types of distributions and their roles in inference.

Comparing Population, Sample, and Sampling Distributions

You'll learn how to distinguish three fundamental types of distributions and understand their distinct roles in statistical inference.

The Three Distribution Types

Think of these like three different photo albums:

Population distribution is the complete album of everyone's heights in a country — every single person. It shows how the entire population is distributed across values. This is what we usually want to know about but rarely can observe directly.

Sample distribution is one snapshot from that album — maybe 100 randomly selected people. It shows how those specific individuals in your sample are distributed. Each time you take a new sample, you get a different snapshot with a slightly different distribution.

Sampling distribution is an album of averages — imagine taking 1,000 different samples of 100 people each, calculating each sample's average height, and plotting those 1,000 averages. This distribution shows how the sample statistic itself (like the mean) varies across repeated samples.

Their Roles in Inference

  • Population distribution: The unknown truth we're trying to learn about (parameters like μ, σ)
  • Sample distribution: The data we actually collect (gives us statistics like x̄, s)
  • Sampling distribution: The bridge between them — tells us how much our sample statistic would vary if we repeated the sampling process, enabling us to make inferences about the population

The sampling distribution is what lets us say things like "our sample mean is probably within 2 points of the true population mean" — it quantifies the uncertainty introduced by sampling.

Key Takeaway: Population distributions describe entire groups, sample distributions describe specific datasets, and sampling distributions describe how statistics behave across repeated sampling — this last one is the foundation of statistical inference.