Skip to content
AP Statistics · Inference for Quantitative Data: Means

CLT

CLT appears in 29 reviewed AP Statistics questions and sits in Inference for Quantitative Data: Means. Below are worked examples showing exactly how the exam tests it, with the answer marked and every distractor explained.

How CLT is tested

Example 1Harder
A teacher tells students: 'If you flip a coin 1,000 times, you'll get almost exactly 500 heads.' A statistician disagrees. Which of the following BEST captures the correct statistical understanding?
  1. ABoth are wrong — probability is unpredictable and heads count could be anywhere from 0 to 1,000.
  2. BThe teacher is right — with 1,000 flips, you always get exactly 500 heads.
  3. CThe teacher is right only if the coin is perfectly fair to 10 decimal places.
  4. DThe statistician is right — the Law of Large Numbers says the PROPORTION of heads converges to 0.5, but the actual NUMBER of heads will likely deviate from 500 by some amount (typically by about 16 heads in either direction).Correct

Why D is right

The Law of Large Numbers says the PROPORTION of heads converges to 0.5, not that the count is exactly 500. The expected count of heads is 500, but the standard deviation is sqrt(n*p*(1-p)) = sqrt(250) ≈ 15.8. So the count will typically deviate from 500 by about 16 in either direction. The PROPORTION (count/1000) converges to 0.5, but the absolute count has variability.
Why the other 3 choices miss
A
Incorrect because probability does not make all outcomes equally likely in this context. By the CLT and LLN, heads count concentrates around 500 with decreasing relative variability as n grows. Outcomes far from 500 are extremely rare.
B
Incorrect because even a perfectly fair coin will produce a count that varies around 500. Getting exactly 500 heads in 1,000 flips has probability C(1000,500)*(0.5)^1000 ≈ 0.025 — less than 3% chance.
C
Incorrect because the LLN and CLT apply to any fair coin (probability exactly 0.5), and the convergence is probabilistic, not dependent on measurement precision of the fairness.
Example 2Harder
A distribution has heavy tails (kurtosis much greater than 3). A sample of n = 100 is taken from this distribution. The CLT predicts that the sampling distribution of x-bar will be approximately normal. How much confidence should we have in this approximation compared to a sample from a normal population?
  1. ANo confidence — CLT does not apply to heavy-tailed distributions.
  2. BThe same confidence — CLT always gives exact normality for n = 100.
  3. CLess confidence — heavy-tailed populations require larger samples for the CLT approximation to be accurate.Correct
  4. DMore confidence — heavy tails actually make the CLT more accurate.

Why C is right

While the CLT applies broadly, the speed of convergence to normality depends on the population's distribution. Heavy-tailed distributions (high kurtosis) converge more slowly than distributions close to normal. For heavy-tailed populations, n = 100 may not provide as accurate a normal approximation as for a symmetric, light-tailed distribution. More observations are needed for heavy-tailed cases.
Why the other 3 choices miss
A
Incorrect because the CLT does apply to heavy-tailed distributions (as long as the distribution has finite variance). It just requires a larger sample for a good approximation compared to distributions closer to normal.
B
Incorrect because the CLT provides an approximation, not exact normality, for non-normal populations. The accuracy of the approximation depends on both n and the population's shape. Heavy tails require larger n.
D
Incorrect because heavy tails actually slow down convergence to normality in the sampling distribution. More extreme observations in heavy-tailed distributions introduce more variability in x-bar.
Example 3Medium
For the sampling distribution of p-hat, as sample size n increases (with p held constant), what happens to the shape and spread of the distribution?
  1. AThe shape approaches normal and the spread decreases.Correct
  2. BThe shape becomes more skewed and the spread increases.
  3. CBoth shape and spread remain constant since p is held fixed.
  4. DThe shape remains constant but the spread decreases.

Why A is right

As n increases: (1) the sampling distribution of p-hat approaches normality by the CLT (the conditions np ≥ 10 and n(1-p) ≥ 10 become more easily satisfied), and (2) the standard deviation σ_p-hat = sqrt(p*(1-p)/n) decreases (less spread). Both the shape and spread change beneficially with larger n.
Why the other 3 choices miss
B
Incorrect because larger n makes the sampling distribution MORE symmetric (normal), not more skewed. The spread also decreases, not increases.
C
Incorrect because both the shape (approaches normal) and the spread (decreases, as SD = sqrt(p(1-p)/n) → 0) change as n increases. Holding p constant does not prevent these changes.
D
Incorrect because the shape does change — it becomes more normal as n increases. For small n and extreme p, the distribution can be quite skewed; for large n, it approaches normal.

Studied alongside

Other concepts from Inference for Quantitative Data: Means.