In AP Statistics, a parameter is a single, fixed numerical value that measures a characteristic of a population or the distribution of a random variable (like μ, σ, p, or the true slope β). You usually can't calculate it directly, so you estimate it with a statistic from a sample.
A parameter is a number that describes an entire population or the distribution of a random variable. The CED is specific about this in Topic 4.8: a parameter is a single, fixed value. It doesn't wobble from sample to sample, because it isn't computed from a sample at all. The population mean μ, population standard deviation σ, population proportion p, and the true slope of a population regression model β are all parameters.
Here's the catch that makes all of statistics necessary. You almost never get to see the parameter. Measuring every melon a distributor ships or every customer who orders a water cup isn't realistic, so the true value of μ or p stays hidden. Instead, you take a sample, compute a statistic (x̄, s, p̂, or b), and use it to estimate the parameter. Think of the parameter as the fixed target and the statistic as your best shot at it. Every confidence interval and significance test in Units 6-9 is built around one question: what does my sample statistic tell me about the unseen parameter?
Parameters thread through the entire course. In Unit 1 (LO 1.7.A), you learn that a statistic summarizes sample data, which only makes sense once you know parameters summarize populations. In Unit 4 (LOs 4.8.A and 4.8.B), you calculate and interpret parameters for discrete random variables, like the expected value μ_X = Σxᵢ·P(xᵢ). Unit 5 (LO 5.1.A) flips the logic: statistics vary from sample to sample while the parameter sits still, and that variation is exactly what sampling distributions describe. By Unit 9, the parameter of interest is β, the true population slope, and LOs 9.3.A through 9.5.C all ask you to estimate it with a confidence interval or test a claim about it. If you can't name the parameter, you can't define hypotheses, you can't interpret a p-value, and you can't write a conclusion in context. It's the backbone of every inference procedure on the exam.
Keep studying AP Statistics Unit 4
Statistic (Unit 1)
The statistic is the parameter's sample-based estimate. x̄ estimates μ, p̂ estimates p, b estimates β. The statistic changes with every new sample, while the parameter never moves. This pairing is the single most-tested vocabulary distinction in the course.
Sampling Distribution (Unit 5)
A sampling distribution shows how a statistic varies around the fixed parameter across repeated samples. LO 5.1.A starts here by asking why your sample isn't like someone else's. The parameter is the center the whole distribution is anchored to.
Confidence Interval (Units 6-9)
A confidence interval is a range of plausible values for a parameter. Per LO 9.3.A, saying '95% confidence' means roughly 95% of intervals built this way capture the true population slope. The interval moves from sample to sample, but β does not.
Slope of a Regression Model (Unit 9)
In Unit 9, the parameter is β, the slope of the population regression line. The test statistic t = (b − β)/SE_b (LO 9.5.A) measures how far your sample slope b sits from the hypothesized parameter value, in standard errors.
Multiple choice loves the parameter-versus-statistic distinction. Expect stems that describe a study and ask you to identify the parameter of interest, or that hand you a confidence interval like (0.35, 0.65) for a slope and ask for the correct interpretation. The trap answer always treats the interval as capturing the sample slope b. The interval is about β, the population parameter. Practice questions also test the df detail from LO 9.5.A: a simple linear regression with one parameter (the slope) uses df = n − 1 for the slope test in that framing, so a 12-observation study is a setup for a degrees-of-freedom question.
On FRQs, parameters show up in every inference problem. The 2017, 2018, and 2019 exams all included questions (water-cup proportions, melon diameters, ACL recovery, tumbleweed counts) where defining the parameter in context was step one. Hypotheses must be written about parameters, never statistics. H₀: β = 0 earns credit; H₀: b = 0 does not. And per LO 9.5.B, your p-value interpretation must say it was computed assuming the true population parameter equals the null value.
A parameter describes the population and is a single, fixed (usually unknown) value, like μ, σ, p, or β. A statistic describes a sample and varies from sample to sample, like x̄, s, p̂, or b. Mnemonic: Population goes with Parameter, Sample goes with Statistic. On the exam, hypotheses and confidence interval interpretations must reference the parameter. Writing H₀ about x̄ or saying an interval 'captures the sample mean' loses points, because you already know the statistic exactly. Inference exists for the unknown parameter.
A parameter is a single, fixed numerical value that measures a characteristic of a population or the distribution of a random variable.
Parameters use Greek or population notation (μ, σ, p, β), while their sample estimates, called statistics, use x̄, s, p̂, and b.
You almost never know a parameter's true value, so the whole point of inference is using a sample statistic to estimate or test a claim about it.
Hypotheses are always written about parameters, so H₀: β = 0 is correct and H₀: b = 0 is wrong.
A confidence interval gives a range of plausible values for the parameter, and the confidence level describes how often the method captures the true parameter in repeated sampling.
Statistics vary from sample to sample, but the parameter stays put, which is why sampling distributions in Unit 5 are centered on the parameter.
A parameter is a single, fixed number that describes a characteristic of a population or a random variable's distribution, such as the population mean μ, population proportion p, or true regression slope β. You typically can't observe it directly, so you estimate it from a sample.
A parameter describes the population and is fixed (μ, σ, p, β); a statistic describes a sample and varies from sample to sample (x̄, s, p̂, b). Remember: Population pairs with Parameter, Sample pairs with Statistic.
No. The sample mean x̄ is a statistic because it's computed from sample data and changes with each new sample. The population mean μ is the parameter that x̄ estimates.
Always parameters. For a slope test, H₀: β = 0 is correct, while H₀: b = 0 loses credit because b is a known sample value, and there's nothing to test about a number you already have.
No, and this wording costs points. The parameter is fixed, so any specific interval either captures it or doesn't. The 95% refers to the method: in repeated random sampling, about 95% of intervals built this way capture the true parameter.