Scribe logoScribeNotes & practice
Sign inStart free
  1. Notes
  2. ›AP Statistics
  3. ›3 Inference for Categorical Data: Proportions
  4. ›3.2 Sample proportions

AP Statistics · Unit 3 · Inference for Categorical Data: Proportions

3.2 Sample proportions

Describe the sampling distribution of a sample proportion and check when a normal model fits it.

  • 4 min read
  • 5 parts
  • 9 questions
  • Draft
Start a Quick 5 on this topic→Jump to the 9 questions
1/5A sample proportion turns a count into a fraction
  1. 1A sample proportion turns a count into a fraction
  2. 2The sample proportion is centered at the population proportion
  3. 3Larger samples give less variable sample proportions
  4. 4The shape is approximately normal when counts are large
  5. 5Three checks justify a normal model for the sample proportion
  6. Check your recall
  7. Practice: 9 questions
1

Part 1 of 5: A sample proportion turns a count into a fraction

A bakery tests a new recipe on 200 random customers. Each customer answers yes or no. Of them, 124 say they like it.

The fraction who like it is the sample proportion, written p^\hat{p}p^​. Here p^=124200=0.62\hat{p}=\frac{124}{200}=0.62p^​=200124​=0.62. The fraction of all customers who like the recipe is the population proportion, written ppp.

The value of ppp is fixed but unknown. The value of p^\hat{p}p^​ changes with each random sample. So p^\hat{p}p^​ has its own distribution across samples, which the next section describes. This mirrors the sample mean, but for data with two outcomes.

Key terms · tap one for its definition

sample proportion

The sample proportion p^\hat{p}p^​ is the number of successes in a sample divided by the sample size nnn.

population proportion

The population proportion ppp is the fraction of all individuals in the population that have the characteristic of interest.

Part 1 done

2

Part 2 of 5: The sample proportion is centered at the population proportion

Now imagine many random samples of 200 customers, each giving its own p^\hat{p}p^​. Plot all of those values. The result is the sampling distribution of p^\hat{p}p^​.

Its mean equals the population proportion: μp^=p\mu_{\hat{p}}=pμp^​​=p. Some samples give a p^\hat{p}p^​ above ppp, and others give one below. On average they balance, so estimates do not run high or low.

Like any sampling distribution, it has a center, a spread, and a shape. The next two sections give the spread and the shape.

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 symbol for a sample proportion.
Show answer and explanation

Answer: C · Worked explanation

The symbol p^\hat{p}p^​ stands for the sample proportion. The symbol ppp is the population proportion, which is fixed and usually unknown. The symbol xˉ\bar{x}xˉ is a sample mean, and σp^\sigma_{\hat{p}}σp^​​ is the standard deviation of the sampling distribution of p^\hat{p}p^​, not a single sample value.
3

Part 3 of 5: Larger samples give less variable sample proportions

The spread of p^\hat{p}p^​ depends on the population proportion and the sample size. Its standard deviation is σp^=p(1−p)n\sigma_{\hat{p}}=\sqrt{\frac{p(1-p)}{n}}σp^​​=np(1−p)​​. This formula applies when the sample is no more than 10% of the population.

Because nnn is in the denominator, larger samples give a smaller spread. Quadrupling nnn halves σp^\sigma_{\hat{p}}σp^​​. The product p(1−p)p(1-p)p(1−p) is largest at p=0.5p=0.5p=0.5, so proportions near one half vary the most.

A bigger sample averages out chance variation. The proportions from different samples then stay closer to ppp. Notice that the center does not move with nnn, and only the spread changes.

Figure 1. For a population proportion of 0.40, the larger sample gives a narrower, taller curve with the same center.

Worked example

Mean and standard deviation of a sample proportion

In a large city, 40% of households recycle. Random samples of 100 households are taken. Calculate the mean and standard deviation of p^\hat{p}p^​.
  1. 1
    The mean is μp^=p=0.40\mu_{\hat{p}}=p=0.40μp^​​=p=0.40.
  2. 2
    The standard deviation is σp^=p(1−p)n\sigma_{\hat{p}}=\sqrt{\frac{p(1-p)}{n}}σp^​​=np(1−p)​​.
  3. 3
    Substitute: σp^=0.40(0.60)100=0.0024\sigma_{\hat{p}}=\sqrt{\frac{0.40(0.60)}{100}}=\sqrt{0.0024}σp^​​=1000.40(0.60)​​=0.0024​.
  4. 4
    Evaluate: 0.0024≈0.049\sqrt{0.0024}\approx 0.0490.0024​≈0.049.

Answer: The mean of p^\hat{p}p^​ is 0.40 and its standard deviation is about 0.049.

Part 3 done

4

Part 4 of 5: The shape is approximately normal when counts are large

Center and spread are settled, so shape comes next. A sample proportion comes from counting successes, and the count can be skewed. Take p=0.02p=0.02p=0.02 and n=100n=100n=100. Most samples hold 0 to 4 successes, and large counts are rare.

As the expected counts grow, the skew fades and the distribution becomes approximately normal. The large counts condition checks this. Here npnpnp is the expected number of successes, and n(1−p)n(1-p)n(1−p) is the expected number of failures. Counting both is a quick test of shape.

Common mistake

A sample size of at least 30 is not the test here. A large nnn with a very small ppp can still fail the condition.

Key term

large counts condition

The large counts condition requires that both npnpnp and n(1−p)n(1-p)n(1−p) are at least 10, so the sampling distribution of p^\hat{p}p^​ is approximately normal.

Part 4 done

Quick check

One question on the part you just read

Easy

Identify the mean of the sampling distribution of p^\hat{p}p^​.
Show answer and explanation

Answer: B · Worked explanation

The mean of the sampling distribution of p^\hat{p}p^​ is the population proportion ppp, so sample proportions are centered on ppp. The symbol p^\hat{p}p^​ changes from sample to sample. The value npnpnp is an expected count, and the square root expression is the standard deviation, not the mean.
5

Part 5 of 5: Three checks justify a normal model for the sample proportion

Before using a normal model, confirm three conditions. Write each one with the numbers from the problem.

  • Random: the data come from a random sample.
  • The 10% condition: the sample size nnn is at most 10% of the population size NNN.
  • Large counts: both npnpnp and n(1−p)n(1-p)n(1−p) are at least 10.

Sampling without replacement makes selections slightly dependent. Keeping nnn at most 10% of NNN makes this effect negligible. If any condition fails, do not use the normal model.

Worked example

Checking conditions

A town has 5,000 voters, and 40% favor a measure. A random sample of 150 voters is taken. Check the conditions for a normal model for p^\hat{p}p^​, then describe the model.
  1. 1
    Random: the sample is stated to be random.
  2. 2
    10% condition: 10% of 5,000 is 500, and 150≤500150\le 500150≤500.
  3. 3
    Large counts: np=150(0.40)=60≥10np=150(0.40)=60\ge 10np=150(0.40)=60≥10 and n(1−p)=150(0.60)=90≥10n(1-p)=150(0.60)=90\ge 10n(1−p)=150(0.60)=90≥10.
  4. 4
    Standard deviation: 0.40(0.60)150=0.0016=0.040\sqrt{\frac{0.40(0.60)}{150}}=\sqrt{0.0016}=0.0401500.40(0.60)​​=0.0016​=0.040.

Answer: All conditions are met. p^\hat{p}p^​ is approximately normal with mean 0.40 and standard deviation 0.040.

Exam tip

Write a justification in context and show the numbers. Stating only that the conditions are met earns no credit.

Key term

10% condition

The 10% condition requires that the sample size nnn is at most 10% of the population size NNN when sampling without replacement.

Note read

Check your recall[13 marks]

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

  1. 1.Calculate[3 marks]

    The population proportion is p=0.30p=0.30p=0.30. Calculate the mean and standard deviation of the sampling distribution of p^\hat{p}p^​ for random samples of size 200.

    Show mark schemeHide mark scheme

    Model answer

    The mean is μp^=p=0.30\mu_{\hat{p}}=p=0.30μp^​​=p=0.30. The standard deviation is 0.30(0.70)200=0.00105≈0.0324\sqrt{\frac{0.30(0.70)}{200}}=\sqrt{0.00105}\approx 0.03242000.30(0.70)​​=0.00105​≈0.0324.

    • 1 mark for stating that the mean is 0.30
    • 1 mark for substituting into p(1−p)n\sqrt{\frac{p(1-p)}{n}}np(1−p)​​
    • 1 mark for obtaining a standard deviation of about 0.0324
  2. 2.Describe[2 marks]

    Describe the shape of the sampling distribution of p^\hat{p}p^​ when p=0.50p=0.50p=0.50 and n=100n=100n=100.

    Show mark schemeHide mark scheme

    Model answer

    It is approximately normal. The large counts condition holds, since np=50np=50np=50 and n(1−p)=50n(1-p)=50n(1−p)=50 are both at least 10.

    • 1 mark for stating that the shape is approximately normal
    • 1 mark for supporting this with the large counts condition and the values 50 and 50
  3. 3.Justify[3 marks]

    A random sample of 50 is taken from a very large population in which p=0.06p=0.06p=0.06. Justify whether a normal model for p^\hat{p}p^​ is appropriate.

    Show mark schemeHide mark scheme

    Model answer

    It is not appropriate. Here np=50(0.06)=3np=50(0.06)=3np=50(0.06)=3, which is less than 10, so the large counts condition fails. The distribution is likely skewed to the right.

    • 1 mark for concluding that the normal model is not appropriate
    • 1 mark for calculating np=3np=3np=3
    • 1 mark for stating that 3 is less than 10, so the large counts condition fails
  4. 4.Predict[2 marks]

    The standard deviation of p^\hat{p}p^​ is 0.04 for samples of size 100. Predict its value for samples of size 400 from the same population.

    Show mark schemeHide mark scheme

    Model answer

    It will be 0.02. The sample size is multiplied by 4, so the standard deviation is divided by 4=2\sqrt{4}=24​=2.

    • 1 mark for predicting a standard deviation of 0.02
    • 1 mark for linking this to dividing by the square root of 4
  5. 5.Identify[1 mark]

    Identify the name of the condition that requires np≥10np\ge 10np≥10 and n(1−p)≥10n(1-p)\ge 10n(1−p)≥10.

    Show mark schemeHide mark scheme

    Model answer

    It is the large counts condition.

    • 1 mark for identifying the large counts condition
  6. 6.Explain[2 marks]

    Explain why the 10% condition is checked when a random sample is taken without replacement.

    Show mark schemeHide mark scheme

    Model answer

    Sampling without replacement makes later selections slightly dependent on earlier ones. When the sample is at most 10% of the population, this effect is negligible and the formula for the standard deviation still applies.

    • 1 mark for stating that sampling without replacement makes selections slightly dependent
    • 1 mark for explaining that a sample of at most 10% of the population makes the effect negligible

Practice this topic

9 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 9 answered

Includes the 2 quick checks in the note above.

3.

Easy·Calculate[1 mark]

In a random sample of 60 students, 18 own a bike. Calculate the sample proportion.
Show answer and explanation

Answer: D · Worked explanation

The sample proportion is the count divided by the sample size: p^=1860=0.30\hat{p}=\frac{18}{60}=0.30p^​=6018​=0.30. The value 0.70 is the proportion who do not own a bike. The value 3.33 divides the sample size by the count instead, which inverts the fraction. The value 0.18 is just the count written as a decimal.
4.

Medium·Determine[1 mark]

A random sample of 200 is taken from a population of 50,000 in which p=0.03p=0.03p=0.03. Determine which condition for a normal model is not met.
Show answer and explanation

Answer: D · Worked explanation

The expected number of successes is np=200(0.03)=6np=200(0.03)=6np=200(0.03)=6, which is below 10, so large counts fails. The 10% condition holds because 10% of 50,000 is 5,000, which is far larger than 200. The count of failures, 194, is well above 10.
5.

Medium·Determine[1 mark]

The standard deviation of p^\hat{p}p^​ is 0.06 for random samples of size 100. Determine its value for samples of size 400 from the same population.
Show answer and explanation

Answer: B · Worked explanation

The sample size is multiplied by 4, so the standard deviation is divided by 4=2\sqrt{4}=24​=2, giving 0.062=0.03\frac{0.06}{2}=0.0320.06​=0.03. The value 0.015 divides by 4 and forgets the square root. The value 0.12 doubles the spread, and 0.06 ignores the larger sample.
6.

Medium·Identify[1 mark]

Identify the situation in which the large counts condition is met.
Show answer and explanation

Answer: D · Worked explanation

With p=0.50p=0.50p=0.50 and n=30n=30n=30, np=15np=15np=15 and n(1−p)=15n(1-p)=15n(1−p)=15, so both are at least 10. In the others, one count is too small: 5 failures, 5 successes, and 8 successes. A small sample can meet the condition when ppp is near one half, while a large sample can fail when ppp is tiny.
7.

Hard·Calculate[2 marks]

For random samples of 100 from a population with p=0.40p=0.40p=0.40, the sampling distribution of p^\hat{p}p^​ is approximately normal. Calculate the probability that p^\hat{p}p^​ exceeds 0.45.
Show answer and explanation

Answer: B · Worked explanation

The standard deviation is 0.40(0.60)100≈0.049\sqrt{\frac{0.40(0.60)}{100}}\approx 0.0491000.40(0.60)​​≈0.049. Then z=0.45−0.400.049≈1.02z=\frac{0.45-0.40}{0.049}\approx 1.02z=0.0490.45−0.40​≈1.02, and P(Z>1.02)≈0.154P(Z>1.02)\approx 0.154P(Z>1.02)≈0.154. The value 0.846 is the area below z=1.02z=1.02z=1.02. The value 0.459 comes from forgetting to divide by nnn, which makes the spread far too large. The value 0.023 matches a zzz-score of 2, which this problem does not give.
8.

Hard·Determine[2 marks]

The population proportion stays fixed while the random sample size increases. Determine which statement correctly describes the sampling distribution of p^\hat{p}p^​.
Show answer and explanation

Answer: D · Worked explanation

The mean of p^\hat{p}p^​ is always ppp, whatever the sample size. Since σp^=p(1−p)n\sigma_{\hat{p}}=\sqrt{\frac{p(1-p)}{n}}σp^​​=np(1−p)​​, a larger nnn reduces the spread, and the larger counts make the shape closer to normal. A sample size never moves the mean toward 0.5 or makes it grow.
9.

Hard·Determine[2 marks]

Random samples of size 100 are taken from population A with p=0.50p=0.50p=0.50 and population B with p=0.10p=0.10p=0.10. Determine which sampling distribution of p^\hat{p}p^​ has the larger standard deviation.
Show answer and explanation

Answer: A · Worked explanation

For A, 0.50(0.50)100=0.050\sqrt{\frac{0.50(0.50)}{100}}=0.0501000.50(0.50)​​=0.050. For B, 0.10(0.90)100=0.030\sqrt{\frac{0.10(0.90)}{100}}=0.0301000.10(0.90)​​=0.030. The product p(1−p)p(1-p)p(1−p) is largest at 0.50, so A varies more. Equal sample sizes do not give equal spreads, because the spread also depends on ppp. Rare outcomes do not make a proportion vary more.

Keep going

Next topic3.3 Confidence interval for a proportionPrevious topic3.1 Estimators

More in Unit 3 · Inference for Categorical Data: Proportions

  1. 3.1EstimatorsNotes · 10 questions
  2. 3.2Sample proportionsNotes · 9 questions
  3. 3.3Confidence interval for a proportionNotes · 10 questions
  4. 3.4Claims from a proportion intervalNotes · 10 questions
  5. 3.5Setting up a test for a proportionNotes · 10 questions
  6. 3.6p-valuesNotes · 10 questions
  7. 3.7Carrying out a test for a proportionNotes · 10 questions
  8. 3.8Errors in testsNotes · 10 questions
  9. 3.9Differences in sample proportionsNotes · 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. 1A sample proportion turns a count into a fraction
  2. 2The sample proportion is centered at the population proportion
  3. 3Larger samples give less variable sample proportions
  4. 4The shape is approximately normal when counts are large
  5. 5Three checks justify a normal model for the sample proportion
  6. Check your recall
  7. Practice: 9 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 of a sample proportion

Density0.10.20.30.40.50.60.702468Sample proportion
  • n = 25
  • n = 100