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

T distribution

T distribution appears in 14 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 T distribution is tested

Example 1Medium
A nutritionist measures the daily caloric intake of a random sample of 36 adults and obtains x-bar = 2,150 calories with s = 420 calories. She constructs a 95% confidence interval for the population mean daily caloric intake. Using t* = 2.030 (for df = 35), the confidence interval is
  1. A2,150 ± 1.960(420/sqrt(36)) = 2,150 ± 137.20 = (2,012.80, 2,287.20)
  2. B2,150 ± 2.030(420/sqrt(36)) = 2,150 ± 142.10 = (2,007.90, 2,292.10)Correct
  3. C2,150 ± 2.030(420/36) = 2,150 ± 23.68 = (2,126.32, 2,173.68)
  4. D2,150 ± 2.030(420) = 2,150 ± 852.60 = (1,297.40, 3,002.60)

Why B is right

The confidence interval formula is x-bar ± t*(s/sqrt(n)). SE = s/sqrt(n) = 420/sqrt(36) = 420/6 = 70. Margin of error = t* * SE = 2.030 * 70 = 142.10. CI = 2,150 ± 142.10 = (2,007.90, 2,292.10).
Why the other 3 choices miss
A
Incorrect. This uses z* = 1.960 instead of t* = 2.030. Since the population standard deviation is unknown (using s), the t-distribution should be used.
C
Incorrect. This divides by n (36) instead of sqrt(n) (6). The standard error formula uses sqrt(n) in the denominator.
D
Incorrect. This uses s = 420 directly without dividing by sqrt(n). The standard error accounts for the sample size, reducing the variability of the mean estimate.

Studied alongside

Other concepts from Inference for Quantitative Data: Means.