Inference from sample statistics and margin of error

Official Digital SAT skill — Problem-Solving and Data Analysis domain.

What this question tests

This skill tests whether you can interpret a sample-based estimate and its margin of error as a statement about what is plausible for the population, not as a guarantee and not as a description of just the sample. On the test, you will usually see a survey or poll result like a proportion or mean, along with a margin of error, and you must decide which conclusion fits the uncertainty correctly. The margin of error is there to remind you that a random sample gives an estimate, not the exact population value, and the interval around the estimate is the reasonable range for the population parameter. You are being tested on careful language: “plausibly between” is very different from “exactly” and from statements about only the surveyed group. This matters because real data decisions depend on acknowledging sampling variability, and the test wants to see that you can connect the statistic (from the sample) to the parameter (for the population) appropriately.

You may see answer choices that tempt you to treat the reported estimate as exact, or to apply the margin-of-error interval to the respondents rather than the population, or to make incorrect claims about how the margin of error changes with sample size. The correct choice usually builds the interval using the given estimate and margin of error and then frames the conclusion as a plausible range for the population described by the sampling process. You do not need to compute the margin of error from scratch when it is provided; your job is to interpret it correctly. You also must avoid overreaching beyond the population that was actually sampled, even if the wording sounds broad. In short, the test is checking both your statistical interpretation and your precision with what the data can and cannot support.

What to know

A sample statistic is a number computed from the sample, such as a sample proportion $\hat{p}$ or sample mean $\bar{x}$, while a parameter is the true (usually unknown) value for the population, such as the population proportion $p$ or population mean $\mu$.
A margin of error quantifies typical sampling uncertainty around a sample estimate, so it should be interpreted as creating a plausible range rather than asserting the estimate is exact.
When an estimate and a margin of error are given, the standard interpretation is the interval $\text{estimate} \pm \text{MOE}$, such as $\hat{p} \pm \text{MOE}$ for a proportion or $\bar{x} \pm \text{MOE}$ for a mean.
The interval describes plausible values for the population parameter under the conditions of the sampling design, especially when the sample is described as random and from the population of interest.
Margin of error is tied to sample size: for many common survey estimates, larger samples generally produce smaller margin of error, so increasing $n$ tends to narrow the plausible interval rather than widen it.
A confidence level (often associated with margin of error) reflects the long-run success rate of the interval method, meaning it is not a guarantee that the specific interval from one sample must contain the true parameter.

How to approach it

First, identify the estimate being reported and what it is estimating, because you must distinguish the sample statistic from the population parameter you are being asked to interpret.
Next, compute the plausible interval using $\text{estimate} \pm \text{MOE}$, because this is the key numerical step that turns a single estimate into a range of reasonable population values.
Then, read the population description carefully and match your conclusion to that population only, because applying the conclusion to a different group is an overgeneralization.
After that, choose wording that emphasizes plausibility and uncertainty, because margin of error statements support a range of possible parameter values rather than an exact claim.
Also, reject choices that describe the sample as if it were uncertain in the same way, because the sample result is already observed and the interval is about what the population parameter could be.
Finally, check any statements about sample size and margin of error for directionality, because larger samples generally reduce margin of error and answer choices often reverse this relationship as a trap.

Common traps

Exact-value trap: a choice claims the population value is exactly the sample estimate, which is appealing because it feels precise, but it ignores sampling variability; avoid it by insisting on a range-based conclusion when a margin of error is given.
Sample-only trap: a choice applies the interval to the surveyed respondents, which sounds reasonable but misses the point; avoid it by remembering the interval is about the population parameter, not the already-recorded sample outcomes.
Overreach trap: a choice generalizes beyond the sampled population (for example, from residents of one region to everyone), which is tempting when the topic feels universal; avoid it by anchoring your conclusion to the population named in the prompt.
Reverse-size trap: a choice says increasing sample size increases margin of error, which can sound intuitive if you mix up variability ideas; avoid it by recalling that larger samples generally mean smaller margin of error.
Guarantee trap: a choice treats the interval as certain to contain the true parameter, which is attractive because it removes ambiguity; avoid it by using language of plausibility and confidence rather than absolute certainty.

Tips & shortcuts

Build the interval immediately from the numbers given, because having the range in front of you makes it easier to spot which choices match the correct interpretation.
Underline the population in the problem statement, because many wrong answers fail by applying the conclusion to the wrong group.
Watch for words like “exactly” or “all” in answers, because they often overstate what a sample estimate with margin of error can justify.
If an option talks about the surveyed sample as if it were unknown, treat it with suspicion, because the uncertainty is about the population parameter, not the observed sample result.

Worked example

A city agency randomly sampled households and found a sample mean weekly water use of $180$ gallons, with a margin of error of $12$ gallons. Which value is a plausible mean weekly water use for all households in the city?

A. $193$ gallons ✓ (correct answer)
B. $200$ gallons
C. $170$ gallons
D. $165$ gallons

Why: The margin of error gives a plausible range for the population mean: $180 \pm 12$, which is from $168$ to $192$ gallons. Only $170$ gallons falls within this interval, so the correct answer is B.

Use the Practice Questions for this skill to drill it, then attempt a Timed Practice Test.