Scribe logoScribeNotes & practice
Sign inStart free
  1. Notes
  2. ›AP Statistics
  3. ›2 Probability, Random Variables and Probability Distributions
  4. ›2.12 Sampling distributions and the central limit theorem

AP Statistics · Unit 2 · Probability, Random Variables and Probability Distributions

2.12 Sampling distributions and the central limit theorem

Explain sampling variability, state the central limit theorem, and predict when a sample mean has an approximately normal sampling distribution.

  • 5 min read
  • 5 parts
  • 10 questions
  • Draft
Start a Quick 5 on this topic→Jump to the 10 questions
1/5Sample statistics change from one random sample to the next
  1. 1Sample statistics change from one random sample to the next
  2. 2A sampling distribution shows every value a statistic can take
  3. 3Larger samples make sample means less variable
  4. 4The Central Limit Theorem describes the shape for large samples
  5. 5Population shape and sample size decide whether normality applies
  6. Check your recall
  7. Practice: 10 questions
1

Part 1 of 5: Sample statistics change from one random sample to the next

Suppose a school has 2,000 students. You ask 40 randomly chosen students how many hours they slept last night. Then you repeat the survey with a different random group of 40. The two averages will almost never match.

The average sleep time of all 2,000 students is a parameter, a number that describes a population. The average from one group of 40 is a statistic, a number calculated from a sample. A statistic changes from sample to sample, and this is called sampling variability.

Common mistake

Sampling variability is not a mistake by the researcher. It is the natural result of random selection, even when the sampling is done well.

Key terms · tap one for its definition

parameter

A parameter is a fixed number that describes a whole population, such as the population mean μ\muμ or proportion ppp.

statistic

A statistic is a number calculated from sample data, such as the sample mean xˉ\bar{x}xˉ or sample proportion p^\hat{p}p^​.

sampling variability

Sampling variability is the way the value of a statistic changes from one random sample to another of the same size.

Part 1 done

2

Part 2 of 5: A sampling distribution shows every value a statistic can take

Now imagine repeating the sleep survey thousands of times, each time with a new random sample of 40. Record every sample mean and make a dot plot. The pattern you see is the sampling distribution of the sample mean.

This distribution shows how far a sample mean is likely to land from the parameter. It has a center, a spread, and a shape.

Three facts describe it when samples are random and each has size nnn. Here μ\muμ and σ\sigmaσ are the population mean and standard deviation.

  • The mean of the sampling distribution is μxˉ=μ\mu_{\bar{x}}=\muμxˉ​=μ.
  • The standard deviation is σxˉ=σn\sigma_{\bar{x}}=\frac{\sigma}{\sqrt{n}}σxˉ​=n​σ​.
  • The second formula needs nnn to be at most 10% of the population.

Key term

sampling distribution

The sampling distribution of a statistic is the distribution of its values over all possible samples of the same size from the population.

Part 2 done

Quick check

One question on the part you just read

Easy

Identify the statement that describes a statistic.
Show answer and explanation

Answer: B · Worked explanation

A statistic is calculated from sample data, so the mean age of 50 chosen employees is a statistic. The other values each describe all of the employees, so they are parameters. A population value is fixed, while a sample value changes from one sample to the next.
3

Part 3 of 5: Larger samples make sample means less variable

Now that the formula is on the table, look at how nnn enters it. The sample size sits under a square root in the denominator. Increasing nnn shrinks σxˉ\sigma_{\bar{x}}σxˉ​, so sample means cluster more tightly around μ\muμ.

Figure 1. For a normal population with mean 50 and standard deviation 10, the larger sample gives a taller, narrower curve around the same center.

To cut the standard deviation of xˉ\bar{x}xˉ in half, you must quadruple the sample size. A bigger sample does not change the population. It changes how far a sample mean can wander from μ\muμ, because extreme values are averaged out.

Worked example

Standard deviation of a sample mean

Scores in a large population have mean 68 and standard deviation 12. Find the mean and standard deviation of the sampling distribution of xˉ\bar{x}xˉ for random samples of size 36 and of size 144.
  1. 1
    The mean is μxˉ=μ=68\mu_{\bar{x}}=\mu=68μxˉ​=μ=68 for both sample sizes.
  2. 2
    For n=36n=36n=36: σxˉ=1236=126=2\sigma_{\bar{x}}=\frac{12}{\sqrt{36}}=\frac{12}{6}=2σxˉ​=36​12​=612​=2.
  3. 3
    For n=144n=144n=144: σxˉ=12144=1212=1\sigma_{\bar{x}}=\frac{12}{\sqrt{144}}=\frac{12}{12}=1σxˉ​=144​12​=1212​=1.
  4. 4
    Quadrupling the sample size halved the standard deviation, from 2 to 1.

Answer: Both distributions have mean 68. The standard deviations are 2 for n=36n=36n=36 and 1 for n=144n=144n=144.

Common mistake

A larger sample does not make σ\sigmaσ smaller. The population spread stays the same, and only the spread of xˉ\bar{x}xˉ shrinks.

Part 3 done

4

Part 4 of 5: The Central Limit Theorem describes the shape for large samples

So far the shape has not come up. Real populations are often skewed, such as clinic waiting times. A few very long waits stretch the right tail.

In plain language, averages of large random samples form a bell shape, even when single values do not. The Central Limit Theorem (CLT) states this for the sampling distribution of xˉ\bar{x}xˉ whatever the population shape.

The center is still μ\muμ and the spread is still σn\frac{\sigma}{\sqrt{n}}n​σ​. A common rule of thumb says n≥30n\ge 30n≥30 is large enough. A larger nnn gives a closer match to the normal curve.

Key term

Central Limit Theorem

The Central Limit Theorem states that the sampling distribution of xˉ\bar{x}xˉ is approximately normal for large random samples, with mean μ\muμ and standard deviation σn\frac{\sigma}{\sqrt{n}}n​σ​.

Part 4 done

Quick check

One question on the part you just read

Easy

Identify what is meant by sampling variability.
Show answer and explanation

Answer: C · Worked explanation

Sampling variability is the change in a statistic from one random sample to the next, even when the methods are careful. Differences among individuals describe the spread of the population, not of the statistic. Favoring some individuals is a flaw in the sampling method, and recording errors are measurement problems.
5

Part 5 of 5: Population shape and sample size decide whether normality applies

Before using a normal model for xˉ\bar{x}xˉ, check the population shape and the sample size. The table sums up the three cases.

When the sampling distribution of xˉ\bar{x}xˉ is approximately normal

Population shapeSample sizeSampling distribution of xˉ\bar{x}xˉ
NormalAny nnnNormal
Skewed or unknownn≥30n\ge 30n≥30Approximately normal
Skewed or unknownn<30n<30n<30Shape not guaranteed
Population shape
Normal
Sample size
Any nnn
Sampling distribution of xˉ\bar{x}xˉ
Normal
Population shape
Skewed or unknown
Sample size
n≥30n\ge 30n≥30
Sampling distribution of xˉ\bar{x}xˉ
Approximately normal
Population shape
Skewed or unknown
Sample size
n<30n<30n<30
Sampling distribution of xˉ\bar{x}xˉ
Shape not guaranteed

Worked example

Checking normality and finding a probability

A right-skewed population of delivery times has mean 20 minutes and standard deviation 8 minutes. Random samples of 64 deliveries are taken. Describe the sampling distribution of xˉ\bar{x}xˉ and find the probability that xˉ\bar{x}xˉ exceeds 22 minutes.
  1. 1
    Since n=64≥30n=64\ge 30n=64≥30, the Central Limit Theorem makes the distribution approximately normal.
  2. 2
    The mean is μxˉ=20\mu_{\bar{x}}=20μxˉ​=20 minutes.
  3. 3
    The standard deviation is σxˉ=864=1\sigma_{\bar{x}}=\frac{8}{\sqrt{64}}=1σxˉ​=64​8​=1 minute.
  4. 4
    Find the zzz-score: z=22−201=2.00z=\frac{22-20}{1}=2.00z=122−20​=2.00.
  5. 5
    Then P(xˉ>22)≈P(Z>2.00)=0.0228P(\bar{x}>22)\approx P(Z>2.00)=0.0228P(xˉ>22)≈P(Z>2.00)=0.0228.

Answer: The distribution is approximately normal with mean 20 minutes and standard deviation 1 minute. The probability is about 0.023.

Exam tip

When you justify normality, name the reason. Either the population is normal, or n≥30n\ge 30n≥30 so the Central Limit Theorem applies.

A small sample from a skewed population keeps some of that skew. A normal model could then mislead you. Also confirm the sample is random and at most 10% of the population. With small samples, plot the data and look for strong skew or outliers.

Go deeperWhy averages look more normal

Individual values can be extreme, but extremes in opposite directions offset each other in an average. As more values are averaged, one outlier carries less weight, and the distribution of the average smooths into a symmetric bell.

The theorem also covers sums and other statistics, such as sample proportions. Very heavy-tailed populations need larger samples than the rule of thumb suggests.

Note read

Check your recall[12 marks]

5 written questions. Answer each one in your head or on paper, then open the mark scheme.

  1. 1.Explain[2 marks]

    Explain why two random samples of the same size from one population usually give different sample means.

    Show mark schemeHide mark scheme

    Model answer

    Each random sample contains different individuals, so the data values differ. A statistic calculated from different data takes a different value, which is sampling variability.

    • 1 mark for stating that different random samples contain different individuals or values
    • 1 mark for linking this to a statistic changing from sample to sample (sampling variability)
  2. 2.Describe[2 marks]

    Describe what the sampling distribution of a sample mean shows.

    Show mark schemeHide mark scheme

    Model answer

    It shows the values of xˉ\bar{x}xˉ over all possible random samples of the same size. It displays the center, spread and shape of how sample means vary.

    • 1 mark for stating that it is the distribution of sample means over all possible samples of the same size
    • 1 mark for stating that it shows how the sample means vary (center, spread or shape)
  3. 3.Calculate[3 marks]

    A population has mean 100 and standard deviation 15. Calculate the mean and standard deviation of the sampling distribution of xˉ\bar{x}xˉ for random samples of size 25.

    Show mark schemeHide mark scheme

    Model answer

    The mean is μxˉ=μ=100\mu_{\bar{x}}=\mu=100μxˉ​=μ=100. The standard deviation is σxˉ=1525=155=3\sigma_{\bar{x}}=\frac{15}{\sqrt{25}}=\frac{15}{5}=3σxˉ​=25​15​=515​=3.

    • 1 mark for stating that the mean of the sampling distribution is 100
    • 1 mark for substituting into σxˉ=σn\sigma_{\bar{x}}=\frac{\sigma}{\sqrt{n}}σxˉ​=n​σ​
    • 1 mark for obtaining a standard deviation of 3
  4. 4.Justify[3 marks]

    A population of household incomes is strongly right skewed. Justify whether the sampling distribution of xˉ\bar{x}xˉ is approximately normal for random samples of size 50.

    Show mark schemeHide mark scheme

    Model answer

    It is approximately normal. Since n=50≥30n=50\ge 30n=50≥30, the Central Limit Theorem applies, so the skewed shape of the population does not prevent a normal sampling distribution.

    • 1 mark for concluding that the sampling distribution is approximately normal
    • 1 mark for citing the Central Limit Theorem or the sample size n≥30n\ge 30n≥30
    • 1 mark for noting that the skewed population shape does not stop this
  5. 5.Predict[2 marks]

    A random sample size is increased from 25 to 100. Predict the effect on the standard deviation of the sampling distribution of xˉ\bar{x}xˉ.

    Show mark schemeHide mark scheme

    Model answer

    The standard deviation is cut in half. The sample size is multiplied by 4, and σxˉ\sigma_{\bar{x}}σxˉ​ is divided by 4=2\sqrt{4}=24​=2.

    • 1 mark for predicting that the standard deviation halves
    • 1 mark for linking this to σn\frac{\sigma}{\sqrt{n}}n​σ​ with nnn multiplied by 4

Practice this topic

10 quick-fire multiple-choice questions, easy to hard. Tap an answer to see if you're right and the worked explanation. (The written questions with mark schemes are under “Check your recall” above.)

0 right0 of 10 answered

Includes the 2 quick checks in the note above.

3.

Easy·Identify[1 mark]

Identify the statement that is true by the Central Limit Theorem for large random samples.
Show answer and explanation

Answer: A · Worked explanation

The Central Limit Theorem is about the distribution of sample means, not about the data. A skewed population stays skewed, and one large sample of data keeps the shape of the population. The mean of the sampling distribution equals μ\muμ, but any single xˉ\bar{x}xˉ still differs from it.
4.

Medium·Calculate[1 mark]

A population has mean 50 and standard deviation 12. Calculate the standard deviation of xˉ\bar{x}xˉ for random samples of size 36.
Show answer and explanation

Answer: B · Worked explanation

Use σxˉ=σn=1236=126=2\sigma_{\bar{x}}=\frac{\sigma}{\sqrt{n}}=\frac{12}{\sqrt{36}}=\frac{12}{6}=2σxˉ​=n​σ​=36​12​=612​=2. The value 0.33 comes from dividing 12 by n=36n=36n=36 instead of by n\sqrt{n}n​. The value 12 is the population standard deviation, which does not change with sample size. The key step is dividing by the square root of the sample size, not by the sample size itself.
5.

Medium·Predict[1 mark]

A right-skewed population is sampled many times with random samples. Predict how the sampling distribution of xˉ\bar{x}xˉ for n=100n=100n=100 compares with that for n=10n=10n=10.
Show answer and explanation

Answer: B · Worked explanation

A larger sample makes σxˉ=σn\sigma_{\bar{x}}=\frac{\sigma}{\sqrt{n}}σxˉ​=n​σ​ smaller, so the distribution is narrower. The Central Limit Theorem also makes it closer to normal for n=100n=100n=100. The mean stays at μ\muμ for both sizes, so averaging more values does not raise it. Both sampling distributions also cluster around the same center.
6.

Medium·Determine[1 mark]

A sampling distribution of xˉ\bar{x}xˉ has standard deviation 6 when n=40n=40n=40. Determine its standard deviation when n=160n=160n=160 for the same population.
Show answer and explanation

Answer: C · Worked explanation

The sample size is multiplied by 4, so the standard deviation is divided by 4=2\sqrt{4}=24​=2. That gives 62=3\frac{6}{2}=326​=3. The value 1.5 would divide by 4 and forget the square root. The value 12 would double the spread, which is the wrong direction.
7.

Medium·Identify[1 mark]

Each situation uses a random sample far smaller than 10% of its population. Identify the situation in which a normal model for xˉ\bar{x}xˉ is reasonable.
Show answer and explanation

Answer: C · Worked explanation

With n=45≥30n=45\ge 30n=45≥30, the Central Limit Theorem makes the sampling distribution approximately normal even though the population is skewed. The other situations have small samples drawn from skewed or unknown populations, so the sampling distribution may keep its skew or outliers.
8.

Hard·Determine[2 marks]

A student says, “With n=100n=100n=100, the population of household incomes becomes approximately normal.” Determine the correct response.
Show answer and explanation

Answer: A · Worked explanation

Taking a sample never changes the population. The Central Limit Theorem describes the distribution of xˉ\bar{x}xˉ over many samples. It applies to skewed populations when nnn is large, so the claim that it needs a normal population is false. No need exists for nnn to equal the population size.
9.

Hard·Calculate[2 marks]

A population has mean 200 and standard deviation 40. For random samples of size 64, the sampling distribution of xˉ\bar{x}xˉ is approximately normal. Calculate the probability that xˉ\bar{x}xˉ exceeds 205.
Show answer and explanation

Answer: A · Worked explanation

Find σxˉ=4064=5\sigma_{\bar{x}}=\frac{40}{\sqrt{64}}=5σxˉ​=64​40​=5. Then z=205−2005=1.00z=\frac{205-200}{5}=1.00z=5205−200​=1.00, and P(Z>1.00)≈0.159P(Z>1.00)\approx 0.159P(Z>1.00)≈0.159. The value 0.450 comes from using the population standard deviation 40, giving z=0.125z=0.125z=0.125. The value 0.841 is the area below z=1.00z=1.00z=1.00, not above it. Always divide the population standard deviation by n\sqrt{n}n​ before finding the zzz-score.
10.

Hard·Determine[2 marks]

Random samples come from a normal population with mean 70 and standard deviation 10. One group takes samples of size 25 and another takes samples of size 100. Determine which group is more likely to get a sample mean above 73.
Show answer and explanation

Answer: A · Worked explanation

For n=25n=25n=25, σxˉ=2\sigma_{\bar{x}}=2σxˉ​=2 and z=1.5z=1.5z=1.5, so the probability is about 0.067. For n=100n=100n=100, σxˉ=1\sigma_{\bar{x}}=1σxˉ​=1 and z=3.0z=3.0z=3.0, so it is about 0.001. Larger samples have a smaller standard deviation, which makes distant sample means less likely, not more likely. The same population and the same cutoff of 73 give very different probabilities.

Keep going

Next topic3.1 EstimatorsPrevious topic2.11 The normal distribution

More in Unit 2 · Probability, Random Variables and Probability Distributions

  1. 2.4Introduction to probabilityNotes · 10 questions
  2. 2.5Mutually exclusive eventsNotes · 10 questions
  3. 2.6Conditional probabilityNotes · 10 questions
  4. 2.7Independent events and unionsNotes · 10 questions
  5. 2.8Random variables and distributionsNotes · 10 questions
  6. 2.9Parameters of random variablesNotes · 10 questions
  7. 2.10Binomial distributionNotes · 10 questions
  8. 2.11The normal distributionNotes · 10 questions
  9. 2.12Sampling distributions and the central limit theoremNotes · 10 questions

All AP Statistics topics →

Draft: written by Scribe and checked automatically, not yet by a subject reviewer.

Practice questions are original, written by Scribe and checked automatically for correctness.

Topic codes follow the official numbering. Not endorsed by College Board or any exam board.

On this page

0/5 read

  1. 1Sample statistics change from one random sample to the next
  2. 2A sampling distribution shows every value a statistic can take
  3. 3Larger samples make sample means less variable
  4. 4The Central Limit Theorem describes the shape for large samples
  5. 5Population shape and sample size decide whether normality applies
  6. Check your recall
  7. Practice: 10 questions
All coursesScribeHow it worksPricingPrivacyTerms

Notes and questions are written by Scribe in our own words. Topic codes follow the official course numbering so you can find your place. Not produced or endorsed by the IB, College Board, AQA, OCR, Pearson Edexcel or any exam board.

© 2026 Scribe

Sampling distributions for two sample sizes

Density3540455055606500.050.10.150.2Sample mean (units of the variable)
  • n = 4 (SD 5)
  • n = 25 (SD 2)