Inference from Sample Statistics and Margin of Error

Inference from sample statistics and margin of error is a focused topic on the Digital SAT that tests whether you understand how data from a sample can be used to draw conclusions about a larger population. You'll need to know what sample mean and sample proportion tell you, how to use them as estimates for population parameters, and what margin of error means in context. Expect to see 1-3 questions on this topic across the math sections. The questions are usually medium difficulty and almost always involve interpreting a survey or study result rather than heavy computation.

Samples vs. Populations

The core idea behind this topic is straightforward: researchers almost never collect data from every single member of a group they're studying. Instead, they collect data from a smaller group (a sample) and use it to make conclusions about the entire group (the population).

• Population parameters are the true values for the entire group. For example, the actual average height of all 10,000 students at a university.
• Sample statistics are the values calculated from the sample. For example, the average height of 200 randomly selected students from that university.

On the SAT, you won't be asked to calculate these from raw data. Instead, you'll be given sample results and asked what they tell you about the population. The key principle: a well-collected sample statistic is a reasonable estimation of the corresponding population parameter, but it's not exact.

Sample mean estimates population mean. If a random sample of 150 customers at a restaurant spent an average of $23.50 per meal, then $23.50 is a reasonable estimate for the average spending of all customers at that restaurant.

Sample proportion estimates population proportion. If 120 out of 400 randomly surveyed voters support a policy (that's 30%), then approximately 30% of all voters in that population likely support the policy.

more resources to help you study

practice questions

Margin of Error

No sample gives you a perfect estimate. The margin of error tells you how far off the sample statistic might be from the true population parameter. It creates a range of plausible values.

If a survey finds that 45% of students prefer online classes with a margin of error of $\pm 4$ percentage points, that means the true population proportion likely falls between:

$45\% - 4\% = 41\%$

$45\% + 4\% = 49\%$

So the true percentage of all students who prefer online classes is plausibly between 41% and 49%.

How Sample Size Affects Margin of Error

This relationship is tested directly on the SAT: a larger sample size generally leads to a smaller margin of error. A bigger sample captures more of the population's variability, so your estimate becomes more precise.

Think of it this way: if you survey 20 people about their favorite food, you might get a quirky result. Survey 2,000 people, and your result will be much closer to what the whole population actually thinks.

You won't need to calculate margin of error on the SAT, but you do need to understand this inverse relationship between sample size and margin of error.

Worked Example 1 (Medium)

A survey of 200 students found that 45% prefer online classes and 55% prefer in-person classes. If the margin of error is $\pm 4$ percentage points, which of the following is a valid conclusion?

Step 1: Find the range for each proportion.

Online preference: $45\% \pm 4\% = 41\%$ to $49\%$

In-person preference: $55\% \pm 4\% = 51\%$ to $59\%$

Step 2: Check the answer choices against these ranges.

The statement "It is plausible that the percentage of all students who prefer in-person classes is between 51% and 59%" falls entirely within the margin of error range for in-person preference. That's a valid conclusion.

A statement like "Exactly 55% of all students prefer in-person classes" would be invalid because the margin of error means we can't pin down the exact value.

Worked Example 2 (Medium)

A researcher surveys a random sample of 500 employees at a large company and finds the sample mean commute time is 34 minutes. The margin of error for this estimate is $\pm 3$ minutes. Which of the following is the best interpretation?

A) All employees commute between 31 and 37 minutes. B) The mean commute time for all employees is plausibly between 31 and 37 minutes. C) Exactly 34 minutes is the mean commute time for all employees. D) Most employees commute exactly 34 minutes.

Solution: The sample mean of 34 minutes estimates the population mean. The margin of error gives a range:

$34 - 3 = 31 \text{ minutes}$

$34 + 3 = 37 \text{ minutes}$

This range applies to the population mean, not to individual employees. Choice B is correct. Choice A is wrong because individual commute times vary widely. Choices C and D treat the estimate as exact, which ignores the margin of error entirely.

Worked Example 3 (Harder)

Two polls estimate the proportion of city residents who support a new park. Poll A surveyed 300 residents and found 62% support with a margin of error of $\pm 5$ percentage points. Poll B surveyed 1,200 residents and found 58% support with a margin of error of $\pm 3$ percentage points. Which statement is supported by these results?

Step 1: Find each poll's range.

Poll A: $62\% \pm 5\% = 57\%$ to $67\%$

Poll B: $58\% \pm 3\% = 55\%$ to $61\%$

Step 2: Notice that the ranges overlap (57% to 61% is common to both). This means the polls are not necessarily contradictory. Also notice that Poll B has a smaller margin of error because it has a larger sample size, making it a more precise estimation.

A valid conclusion: "It is plausible that the true proportion of residents who support the park is between 57% and 61%," since both polls' ranges include those values.

Common Traps and Misinterpretations

Trap 1: Confusing individual values with the population parameter. The margin of error applies to the population mean or population proportion, not to individual data points. If the average commute is $34 \pm 3$ minutes, that doesn't mean every person commutes between 31 and 37 minutes.

Trap 2: Treating the sample statistic as exact. The whole point of margin of error is that the sample result is an estimate. Any answer choice that says the population value is exactly the sample value is almost certainly wrong.

Trap 3: Thinking a larger margin of error is better. A smaller margin of error means a more precise estimate. The SAT may ask which study design would reduce the margin of error, and the answer involves increasing the sample size.

Trap 4: Generalizing beyond the sampled population. If a study surveys college students at one university, the results estimate parameters for students at that university, not all college students everywhere.

What to Watch For on Test Day

1. Read what population was actually sampled. The conclusion can only apply to the group the sample was drawn from. A sample of dog owners in Texas doesn't tell you about all pet owners nationwide.

2. Margin of error creates a range, not a single value. Whenever you see $\pm$ something, immediately calculate the lower and upper bounds. Then check which answer choices fall within that range.

3. Bigger sample = smaller margin of error. If a question asks how to make an estimate more precise, increasing the sample size is the answer. You don't need to memorize a formula for this.

4. Look for the word "plausible." SAT answer choices that use language like "it is plausible that" or "could be" are more likely correct than choices using "definitely" or "exactly," because sample-based estimation always involves uncertainty.

5. Don't overthink the math. These questions are more about reasoning and interpretation than calculation. Most of the time, you're just adding and subtracting the margin of error from the sample statistic and then evaluating statements.