Skip to content
AP Statistics · Probability, Random Variables, and Probability Distributions

Normal approximation

Normal approximation appears in 12 reviewed AP Statistics questions and sits in Probability, Random Variables, and Probability Distributions. Below are worked examples showing exactly how the exam tests it, with the answer marked and every distractor explained.

How Normal approximation is tested

Example 1Medium
A continuous random variable X has a normal distribution. The Central Limit Theorem states that for large samples, the sampling distribution of the sample mean x-bar is approximately normal. This is most useful when:
  1. AThe sample size is exactly 30.
  2. BThe population distribution is unknown or non-normal, allowing inference about the mean using the normal approximation for x-bar.Correct
  3. CThe population standard deviation is known.
  4. DThe population is known to be exactly normal.

Why B is right

The Central Limit Theorem is most powerful when the population distribution is non-normal or unknown. In that case, as long as n is sufficiently large (typically n ≥ 30), the sampling distribution of x-bar is approximately normal, enabling the use of normal-based inference procedures even when the population is not normal.
Why the other 3 choices miss
A
n = 30 is a rule-of-thumb guideline, not a magic threshold. The CLT is useful for large n generally, and 'large' depends on how non-normal the population is.
C
Whether the population SD is known determines which test to use (z vs. t), not whether the CLT applies. The CLT is applicable regardless of whether σ is known.
D
If the population is exactly normal, the sampling distribution of x-bar is normal for ANY sample size, so the CLT's approximation is not needed.
Example 2Harder
In a large population, 30% of adults have Type O blood. A blood bank collects donations from 50 randomly selected adults. Using the CLT, what is the approximate probability that fewer than 10 of the 50 donors have Type O blood?
  1. AApproximately 0.079
  2. BApproximately 0.025
  3. CApproximately 0.159
  4. DApproximately 0.045Correct

Why D is right

Let X be the number of the 50 donors with Type O blood. Each adult independently has a 30% chance of Type O blood, so X is binomial with n = 50 and p = 0.3. The Central Limit Theorem lets us approximate this distribution with a normal curve. Mean: mu = np = 50(0.3) = 15. Standard deviation: sigma = sqrt(npq) = sqrt(50 x 0.3 x 0.7) = sqrt(10.5) = 3.24. We want P(X < 10), i.e. P(X <= 9). Because X is a discrete count, apply the continuity correction and use 9.5 as the boundary: z = (9.5 - 15)/3.24 = -1.70. From the standard normal table, P(Z < -1.70) = 0.045, choice D. The probability is small because 10 donors is well below the expected 15. (For reference, the exact binomial probability is about 0.040, confirming the approximation.)
Why the other 3 choices miss
A
Incorrect. The value 0.079 corresponds to z ≈ −1.41, which does not match the standardized count for fewer than 10 donors.
B
Incorrect. The value 0.025 corresponds to z ≈ −1.96, a cutoff farther into the lower tail than the correct standardized value.
C
Incorrect. The value 0.159 corresponds to z = −1.0, which would indicate about 11.8 donors rather than the boundary near 10.
Example 3Harder
A binomial random variable has n = 20 and p = 0.5. Which of the following is closest to P(X >= 15)?
  1. A0.006
  2. B0.500
  3. C0.250
  4. D0.021Correct

Why D is right

Using the normal approximation: mean = 10, SD = sqrt(20 × 0.5 × 0.5) = sqrt(5) ≈ 2.236. z = (14.5 - 10)/2.236 ≈ 2.01 (using continuity correction). P(Z > 2.01) ≈ 0.022 ≈ 0.021. This can also be computed exactly from the binomial: P(X >= 15) = sum from 15 to 20 ≈ 0.0207.
Why the other 3 choices miss
A
Incorrect because 0.006 corresponds to a more extreme outcome (approximately P(X >= 16) or P(X >= 17)).
B
Incorrect because 0.50 corresponds to P(X >= 10), the median of the distribution.
C
Incorrect because 0.25 would correspond to a much more likely outcome, not one in the extreme upper tail.

Studied alongside

Other concepts from Probability, Random Variables, and Probability Distributions.