Two samples taken from the same population usually give different statistics, and that is called sampling variability. Variation in sample results can come from random chance or from something non-random, like a flawed sampling method, which is why conclusions from a single sample always carry uncertainty.
Why This Matters for the AP Statistics Exam
This is the starting point for all of Unit 5, which carries real weight on the AP Statistics exam. Before you can work with sampling distributions, the Central Limit Theorem, or any inference later in the course, you need to understand why one sample does not perfectly match another or the population.
On the exam you will be expected to:
- Recognize that different samples from the same population produce different statistics.
- Ask the right questions when you see variation, like whether the difference is just random or points to a real problem.
- Tell the difference between a population parameter and a sample statistic, since later procedures depend on knowing which is which.
You will see this thinking show up in multiple-choice questions and as the foundation for free-response questions that ask you to interpret sample results and judge how trustworthy a conclusion is.
Key Takeaways
- A sample statistic almost never equals the population parameter exactly, and two samples of the same size usually differ from each other too.
- Variation between samples from the same population can be random (chance) or non-random (a problem like biased sampling).
- A population parameter is a fixed true value; a sample statistic is a calculated estimate of it.
- Because variation exists, any conclusion drawn from a sample carries some uncertainty.
- Learn the symbols early: parameters and statistics use different notation, and mixing them up causes errors all through the course.
Sampling Variability: Why Your Sample Is Not Like Mine
Suppose you want to estimate the mean income of a group of workers. You take a sample, calculate the sample mean, and get a number. A classmate takes a different sample of the same size and gets a slightly different number. Neither one exactly equals the true population mean. That gap, and the differences between samples, is sampling variability.
The key question this topic asks: when two samples from the same population give different results, why are they different?
There are two possibilities:
- Random variation: just chance. Different random samples naturally include different people, so the statistics wobble a bit around the true value.
- Non-random variation: something systematic, like a biased sampling method, that pushes results in a particular direction.
Spotting the difference matters because random variation is expected and predictable, while non-random variation can make your conclusions wrong.
Parameters vs. Statistics
A population parameter is a measure of a characteristic of a whole population, like the mean or a proportion. It is a fixed value that represents the true value for the population.
A sample statistic is a measure of a characteristic of a sample. It is a calculated value that estimates the population parameter.
A quick memory trick: Statistics come from Samples, and Parameters come from Populations. Here is a chart for the symbols you need:
| Measurement | Population Parameter | Sample Statistic |
|---|---|---|
| Mean | μ | x̅ |
| Standard Deviation | σ | s |
| Proportion | p | p̂ |
Getting comfortable with these symbols now saves you from a lot of confusion when you reach sampling distributions and inference.
Looking Ahead: Sampling Distributions
A sampling distribution is the distribution of a statistic (like the sample mean or sample proportion) across all possible samples of a given size from a population. Instead of plotting individual data values, you plot the statistic from each possible sample.
In earlier units, every distribution came from one sample, such as the grades in one class. A sampling distribution shifts the focus: each data point is the statistic from a different possible sample. This idea is the bridge to the rest of Unit 5, where you will learn how these distributions behave and why they are often approximately normal. For now, the main job in this topic is just recognizing that variation exists and asking why.
How to Use This on the AP Statistics Exam
MCQ
- When a question shows two samples from the same population with different statistics, expect to identify that this is normal sampling variability, not necessarily an error.
- Watch for questions that ask you to decide whether a difference looks random or points to a non-random cause like a biased method.
- Be ready to label a given number as either a parameter or a statistic based on whether it describes a whole population or just a sample.
Free Response
- Use precise language. Say "sample mean" or "sample proportion" when you mean a statistic, and name the population clearly when you mean a parameter.
- If asked why two sample results differ, explain that samples vary by chance, and only call out bias if the setup gives a reason to.
Common Trap
- Do not treat a sample statistic as if it were the true population value. They are usually close but rarely equal.
- Do not assume every difference between samples signals a problem. Some variation is just random.
Common Misconceptions
- "A good sample should match the population exactly." Even a perfect random sample will differ from the population because of chance. The goal is a close estimate, not an exact match.
- "If two samples give different answers, one of them is wrong." Different results from different samples are expected. Variation between samples does not mean a mistake was made.
- "Parameter and statistic mean the same thing." A parameter is the fixed true value for the population; a statistic is an estimate calculated from a sample. They use different symbols on purpose.
- "All sample variation is caused by bias." Variation can be purely random. Bias is only one possible, non-random source of variation.
Practice Problems
(1) A study is conducted to estimate the proportion of students in a school district who have access to the internet at home. A sample of 1000 students is selected from the school district, and it is found that 750 students have access to the internet at home. Is the proportion of students with internet access at home a parameter or a statistic?
(2) The mean height of all adult males in the United States is 70 inches. Is the mean height of adult males a parameter or a statistic?
(3) A survey is conducted to estimate the proportion of adults in a city who have a college degree. A sample of 500 adults is selected from the city, and it is found that 300 adults have a college degree. Is the proportion of adults with a college degree a parameter or a statistic?
(4) The mean income of all households in the United States is $50,000 per year. Is the mean income of households a parameter or a statistic?
(5) A study is conducted to estimate the proportion of employees in a company who are satisfied with their job. A sample of 200 employees is selected from the company, and it is found that 150 employees are satisfied with their job. Is the proportion of employees who are satisfied with their job a parameter or a statistic?
Answers
(1) The proportion of students with internet access at home is a statistic, because it is calculated from a sample of students and is used to estimate the population parameter (the proportion of all students in the school district with internet access at home).
(2) The mean height of adult males is a parameter, because it is a fixed value that represents the true value of the population (all adult males in the United States).
(3) The proportion of adults with a college degree is a statistic, because it is calculated from a sample of adults and estimates the population parameter (the proportion of all adults in the city with a college degree).
(4) The mean income of households is a parameter, because it is a fixed value that represents the true value of the population (all households in the United States).
(5) The proportion of employees who are satisfied with their job is a statistic, because it is calculated from a sample of employees and estimates the population parameter (the proportion of all employees in the company who are satisfied with their job).
Related AP Statistics Guides
Vocabulary
The following words are mentioned explicitly in the College Board Course and Exam Description for this topic.Term | Definition |
|---|---|
population | The entire group of individuals or items from which a sample is drawn and about which conclusions are to be made. |
sample | A subset of individuals or items selected from a population for the purpose of data collection and analysis. |
statistic | Numerical summaries or measures calculated from sample data, such as mean, median, or standard deviation. |
variation | Differences in data that occur by chance due to the random nature of sampling, rather than from systematic causes. |
Frequently Asked Questions
What is sampling variability?
Sampling variability is the natural change in sample statistics from one random sample to another. Even samples from the same population can give different means, proportions, or other statistics.
Why do two samples from the same population differ?
Two samples can differ because random selection includes different individuals each time. If the sampling method is flawed or groups are not comparable, the difference may also come from non-random variation.
What is the difference between random and non-random variation?
Random variation comes from chance in the sampling process. Non-random variation comes from bias, bad sampling design, measurement issues, or real differences between the groups being compared.
What is the difference between a parameter and a statistic?
A parameter describes a population, while a statistic describes a sample. AP Statistics uses sample statistics to estimate population parameters and reason about uncertainty.
What is a sampling distribution?
A sampling distribution is the distribution of a statistic over many possible samples of the same size from the same population. It helps explain how much sample results should vary by chance.
Why does sampling variability matter for AP Statistics?
Sampling variability is the reason AP Statistics uses probability, confidence intervals, and significance tests. It helps you decide whether a sample result is plausible by chance or evidence of something more.