6.3Sampling distribution and the central limit theorem
The idea
The mean of a random sample is itself random. It is centred on the population mean, its spread shrinks like 1/√n, and for large n its shape is approximately normal.
Think of it like this: Ask 100 random people instead of 4: your average answer is far steadier.
Why it exists
This is why polls and measurements work, and the central limit theorem is the reason the normal curve is everywhere.
How it works
- Draw n independent values from a population with mean μ and SD σ.
- The sample mean has mean μ.
- Its standard deviation is σ/√n.
- For large n its distribution is approximately N(μ, σ²/n), whatever the population shape.
- To halve the spread of the sample mean you need four times the data.
Worked example: A skewed population with mean 1 and SD 1; sample size 25
The sample mean has mean 1 and SD 1/√25 = 0.2, and its histogram is close to a bell curve. Compare the histogram bars with the green normal curve in the lab.
Now play with it
- →Start with the exponential (skewed) population and slide n from 1 to 60.
- →Switch to the 0 or 10 population: even this extreme shape becomes a bell for larger n.
Bars = 1500 sample means (each from n draws); green curve = Normal(μ, σ²/n). Whatever the population looks like, the sample mean has mean μ and standard deviation σ/√n, and for larger n its histogram is nearly a bell: the central limit theorem.
Bars = 1500 sample means (each from n draws); green curve = Normal(μ, σ²/n). Whatever the population looks like, the sample mean has mean μ and standard deviation σ/√n, and for larger n its histogram is nearly a bell: the central limit theorem.
Check yourself
A population has SD 12. The sample mean of n = 36 independent draws has standard deviation:
Exam notes (the dense version)
The mean X̄ of n independent draws from a population with mean μ and standard deviation σ has mean μ and standard deviation σ/√n. By the central limit theorem, for large n X̄ is approximately normal, whatever the population looks like. More data narrows the spread by √n.
Previous year questions: Descriptive Statistics & Sampling (2)
PYQs are grouped by topic, so this is the full set for Descriptive Statistics & Sampling; try the ones that match this subtopic. Open all 2 on the topic page.