Skip to content
SAT · Problem Solving and Data Analysis

Margin of error

Margin of error appears in 5 reviewed SAT questions and sits in Problem Solving and Data Analysis. Below are worked examples showing exactly how the exam tests it, with the answer marked and every distractor explained.

How Margin of error is tested

Example 1Harder
A random sample of 100 adults in a city shows 28 own a bicycle. The city has 80,000 adults. Which range best represents the likely true number of bicycle owners, given a margin of error of ±5%?
  1. A18,400 to 22,400
  2. B20,000 to 24,000
  3. C22,400 to 26,400
  4. D18,400 to 26,400Correct

Why D is right

The sample proportion is 28/100 = 28%. With a margin of error of ±5 percentage points, the plausible range for the true proportion is 28% − 5% = 23% to 28% + 5% = 33%. Applying this to the 80,000 adults gives 0.23 × 80,000 = 18,400 and 0.33 × 80,000 = 26,400. So the likely true number of bicycle owners is between 18,400 and 26,400, which is choice D.
Why the other 3 choices miss
A
18,400 to 22,400 is 23% to 28% of 80,000. It applies the margin of error only below the sample estimate and leaves out the upper half of the interval, 28% to 33%.
B
20,000 to 24,000 is 25% to 30% of 80,000. That interval is not centered on the sample proportion of 28%, and it is only 5 percentage points wide instead of the full 10 that ±5% gives.
C
22,400 to 26,400 is 28% to 33% of 80,000. It applies the margin of error only above the sample estimate and leaves out the lower half of the interval, 23% to 28%.
Example 2Medium
A random sample of 100 families shows that 42 have pets. The margin of error is ±5%. Which statement is best supported by these results?
  1. AMost families have pets.
  2. BThe survey is not reliable because the sample is too small.
  3. CBetween 37% and 47% of all families likely have pets.Correct
  4. DExactly 42% of all families have pets.

Why C is right

42% ± 5% = 37% to 47%. This interval represents the likely range for the true population proportion. The data or units point to this result when the values are interpreted correctly.
Why the other 3 choices miss
A
Incorrect: 42% is less than 50%; even the upper bound of 47% is below majority.
B
Incorrect: n = 100 is a reasonable sample size; the margin of error of ±5% indicates sufficient precision.
D
Incorrect: Sample results estimate, not determine exactly, the population proportion; sampling error exists.
Example 3Medium
A survey of 500 people found that the average number of hours of sleep per night is 6.8. The margin of error is ±0.3 hours. Which statement is best supported?
  1. AExactly 500 people sleep 6.8 hours per night.
  2. BThe survey proves people sleep less than 7 hours per night.
  3. CWe can be confident the population average falls between 6.5 and 7.1 hours.Correct
  4. DAll people sleep between 6.5 and 7.1 hours per night.

Why C is right

A margin of error means the estimated interval (6.5, 7.1) likely contains the true population mean. The data or units point to this result when the values are interpreted correctly.
Why the other 3 choices miss
A
Incorrect: 6.8 hours is the sample mean, not the exact sleep time of all 500 respondents.
B
Incorrect: The upper bound is 7.1, so the data doesn't rule out averages of 7 hours or more.
D
Incorrect: The interval applies to the estimated population MEAN, not to every individual's sleep time.

Studied alongside

Other concepts from Problem Solving and Data Analysis.