SAT Sampling, Surveys & Margin of Error
Sampling, Surveys & Margin of Error
From a sample to a population estimate: understand what a sample statistic tells you, how margin of error changes the conclusion, and how the SAT turns these ideas into multi-step problems.
What this lesson covers
- Population vs. sample
- Sample mean and sample proportion
- Using sample statistics to estimate population values
- Margin of error and plausible intervals
- Sample size and margin of error
- Turning estimated percentages into population counts
- Common SAT traps and multi-step reasoning
The SAT focus
The SAT expects students to use a sample mean or sample proportion to estimate a population value, interpret margin of error, and understand that a larger sample size generally leads to a smaller margin of error.
Section 1 — Population vs. Sample
A population is the entire group we want to understand. A sample is a smaller group selected from that population.
Population
The complete group of interest.
Example: All 4,800 students enrolled at a university.
Sample
A subset selected from the population.
Example: 240 students selected from those 4,800 students.
Section 2 — Why Use a Sample?
Studying every member of a large population can be expensive or impractical. Instead, a researcher can collect data from a sample and use a statistic from that sample to estimate a characteristic of the population.
Sample mean
The average of the values in the sample.
It can be used to estimate the population mean, \(\mu\).
Sample proportion
The fraction or percentage of the sample with a particular characteristic.
It can be used to estimate the population proportion, \(p\).
Section 3 — Random Samples and Representing a Population
A sample is useful for estimating a population only when the way the sample is selected supports that goal. A random sample gives members of the population a fair chance of being selected and helps reduce the risk that the sample is systematically different from the population.
Random selection
Each member of the population has a chance to be selected according to the random-sampling method being used.
This helps make the sample more representative of the population of interest.
Potentially biased selection
If a researcher selects only people who are easiest to reach or most likely to volunteer, the sample may not represent the population well.
The detailed study-design consequences of biased sampling belong in Part 8.
Section 4 — Sample Mean as an Estimate of a Population Mean
Suppose a random sample of 40 batteries has a mean lifetime of 18.6 hours. If the goal is to estimate the mean lifetime of all batteries in the population, the sample mean of 18.6 hours is the estimate.
Section 5 — Sample Proportion as an Estimate of a Population Proportion
If 72 of 120 randomly selected customers prefer a new package design, the sample proportion is:
If the sample is being used to estimate the preference of the full population, 60% is the estimate of the population proportion.
Section 6 — From Sample Data to a Population Conclusion
The reasoning can be pictured as a chain:
Once margin of error is given, the SAT often asks you to move one more step: identify the range of plausible population values.
Section 7 — What Margin of Error Means
Margin of error (MOE) describes how far the plausible population value can extend above or below the sample estimate.
For example, if a survey estimates that 46% of the population supports a proposal with a margin of error of 3 percentage points, then the plausible population proportion is:
Section 8 — Margin of Error for Population Means
The same idea works for a sample mean.
So the plausible population mean is between 12.8 and 15.2 hours.
Section 9 — Margin of Error for Population Proportions
For percentages, margin of error is usually expressed in percentage points.
This means the population proportion is plausibly between 48% and 56%.
Section 10 — Margin of Error Applies to the Population Parameter—not Individuals
This is a powerful SAT distinction.
Correct interpretation
If the estimated population mean is 74 with a margin of error of 3, then:
Incorrect interpretation
It does not mean every individual in the population has a value between 71 and 77.
Section 11 — Sample Size and Margin of Error
When the relevant conditions are comparable, a larger random sample generally produces a smaller margin of error.
Smaller sample
Less information about the population → generally more uncertainty.
Larger sample
More information about the population → generally less uncertainty.
Section 12 — Keeping the Population the Same
If a researcher wants a more precise estimate for a particular population, increasing the sample size helps only when the larger sample is still drawn from the same population of interest.
This lesson only needs the basic idea. Detailed questions about biased sampling, generalizability, observational studies, and experiments belong in the next satellite lesson on statistical claims and study design.
Section 13 — From Sample Proportion to Population Count
Sometimes the SAT gives a sample proportion and asks for an estimate of the number of people in the population.
Section 14 — From Margin of Error to a Plausible Population Count
This is a classic multi-step pattern:
- Start with the estimated population proportion.
- Subtract and add the margin of error.
- Convert the two percentages into population counts.
Therefore, a plausible population count is between 6,400 and 7,600 people.
Section 15 — Common SAT Traps
Trap 1: Treating the estimate as exact
An estimate of 48% with MOE 3% does not mean the population proportion is exactly 48%.
Trap 2: Applying MOE to individuals
A mean of 74 ± 3 describes a plausible population mean, not every individual value.
Trap 3: Using the sample size for a population count
If the population has 25,000 people, use 25,000 when converting a population proportion into a population count.
Trap 4: Forgetting what “larger sample” means
A larger sample generally gives a smaller MOE when the comparison concerns the same population and otherwise comparable conditions.
Section 16 — A Reliable SAT Sampling & Margin-of-Error Strategy
- Identify the population. What complete group is the question trying to describe?
- Identify the sample. Which smaller group actually provided the data?
- Identify the statistic. Is the sample giving you a mean or a proportion?
- Use the statistic as the estimate. Sample mean estimates population mean; sample proportion estimates population proportion.
- Apply the margin of error. Find the lower and upper plausible values.
- Keep the target straight. Is the question asking for a population mean, population proportion, or population count?
- For a population count, use the population size. Do not accidentally use the sample size.
- For sample-size comparisons, remember the direction. A larger sample generally means a smaller margin of error.
Advanced Practice
These original questions progress from foundational sample/population reasoning to multi-step margin-of-error interpretation. The visuals are designed to test interpretation rather than decoration.
A university has 2,400 students. A researcher randomly selects 120 students to study their daily study time. Which group is the population in this study?
Answer & Solution
Answer: B. The population is the complete group the researcher wants to understand: all 2,400 students.
A random sample of 40 batteries has a mean lifetime of 18.6 hours. Which value is the best estimate of the mean lifetime of all batteries in the population?
Answer & Solution
Answer: B. The sample mean is 18.6 hours, so it is the estimate of the population mean.
In a random sample of 120 customers, 72 prefer a new package design. What is the sample proportion of customers who prefer the new design?
Answer & Solution
Answer: C. \(72/120=0.60\). The sample proportion is 60%, which can be used to estimate the population proportion.
A random sample of 280 registered voters shows that 84 support a proposal. If the population contains 1,600 registered voters, which is the best estimate of the number of voters in the population who support the proposal?
Answer & Solution
Answer: C. The sample proportion is \(84/280=0.30\). Apply that estimated population proportion to the population size: \(0.30*1,600=480\).
A random sample estimates that 42% of a population has a certain characteristic. The margin of error is 3 percentage points. Which interval represents the plausible values for the population proportion?
Answer & Solution
Answer: A. Subtract and add the margin of error: \(42\%-3\%=39\%\) and \(42\%+3\%=45\%\).
A random sample gives an estimated population mean of 52.6, with a margin of error of 1.4. Which interval represents the plausible values for the population mean?
Answer & Solution
Answer: A. \(52.6-1.4=51.2\) and \(52.6+1.4=54.0\).
A random sample is used to estimate the mean number of minutes students spend commuting to school. The estimated mean is 74 minutes, with a margin of error of 3 minutes. Which statement is supported by this information?
Answer & Solution
Answer: B. The margin of error gives a plausible interval for the population mean: \(74-3=71\) and \(74+3=77\). It says nothing about every individual commute.
Two random samples are taken from the same population under otherwise comparable conditions. Sample A contains 100 people, and sample B contains 400 people. Which sample would generally be expected to have the smaller margin of error?
Answer & Solution
Answer: B. For comparable random samples from the same population, a larger sample generally produces a smaller margin of error.
A researcher randomly selects 150 engineering students to estimate their average weekly study time. The researcher wants to repeat the study with a smaller margin of error while studying the same population. Which change is most likely to help?
Answer & Solution
Answer: B. Increasing the random sample while keeping the population of interest the same generally reduces the margin of error.
A random sample estimates that 41% of a population supports a proposal. The margin of error is 2 percentage points. The population contains 20,000 people. Which interval gives the plausible number of people in the population who support the proposal?
Answer & Solution
Answer: A. First find the plausible population proportions: \(41\%-2\%=39\%\) and \(41\%+2\%=43\%\). Then use the population size: \(0.39*20,000=7,800\) and \(0.43*20,000=8,600\).
A survey reports that the estimated proportion of a population with a certain characteristic is 30%. The plausible interval for the population proportion is 27% to 33%. What is the margin of error?
Answer & Solution
Answer: B. The estimate, 30%, is the midpoint of the interval. The distance from 30% to either endpoint is 3 percentage points.
A random sample from a population of 12,500 people estimates that 62% have used a particular service. The margin of error is 4 percentage points. Which interval gives the plausible number of people in the population who have used the service?
Answer & Solution
Answer: B. The plausible proportions are 58% and 66%. Then \(0.58*12,500=7,250\) and \(0.66*12,500=8,250\).
Two independent surveys estimate the proportion of residents who support a proposed transit plan.
| Survey | Estimate | Margin of Error | Plausible Interval |
|---|---|---|---|
| A | 48% | 2 percentage points | 46% to 50% |
| B | 52% | 5 percentage points | 47% to 57% |
Which statement is supported by the intervals shown?
Answer & Solution
Answer: C. Survey A allows 46%–50%, while survey B allows 47%–57%. The intervals overlap, and 49% lies in both intervals.
The following diagram shows three survey estimates and their margins of error. Which survey has the largest margin of error?
Answer & Solution
Answer: B. Survey B spans the widest interval, so its margin of error is largest. Its interval is 49% to 57%, an 8-point total width, corresponding to a 4-point margin of error.
A random sample of 400 people is selected from a population of 15,000 people. The sample estimates that 34% of the population has a certain characteristic, with a margin of error of 3 percentage points. Which of the following is a plausible number of people in the population with that characteristic?
Answer & Solution
Answer: C. The plausible population proportions are \(34\%-3\%=31\%\) and \(34\%+3\%=37\%\). Therefore the plausible population count is between \(0.31*15,000=4,650\) and \(0.37*15,000=5,550\). The value 5,250 lies within this interval.
New SAT Pattern to Remember
A particularly important multi-step pattern combines three ideas:
When you see this structure, do the work in that order. First find the plausible population proportion. Then apply that proportion to the population size.
One thought on “SAT Sampling, Surveys & Margin of Error”