In AP Statistics, a population is the complete set of all items or subjects of interest in a study. You almost never measure the whole population. Instead, you take a sample from it, calculate a statistic, and use inference to estimate the population's parameters.
A population is everyone (or everything) you actually want to know about. All registered voters in a town. Every steel cable a factory produces. All students at a high school. The CED puts it plainly in Topic 3.2: a population consists of all items or subjects of interest, and a sample is a subset of that population.
Here's the mental model that makes AP Stats click. The population is the thing you care about but usually can't measure directly, because it's too big, too expensive, or literally infinite. So the whole subject is built around a workaround. You take a sample, compute a statistic from it (like x̄ or p̂), and use that statistic to estimate a parameter of the population (like μ or p). Parameters describe populations and are usually unknown. Statistics describe samples and are what you actually have. Every inference procedure in Units 6 through 9 is just a formal way of reasoning from sample back to population.
Population shows up in nearly every unit because it defines what your conclusions are about. In Unit 3, learning objective AP Stats 3.2.A requires you to identify the population and sample in a study, and 3.2.B says you can only generalize to the population your sample actually came from (and only if that sample was randomly selected or otherwise representative). In Unit 5, AP Stats 5.7.A and 5.7.B tie the sampling distribution of x̄ directly to the population's mean μ, standard deviation σ, and shape. If the population is normal, x̄ is normal; if not, you need n ≥ 30. In Unit 8, AP Stats 8.3.D asks you to justify a claim about the population that was sampled based on a chi-square test. Notice the pattern. Defining the population correctly is what separates a full-credit conclusion from a generic one.
Keep studying AP Statistics Unit 6
Sample (Unit 3)
The sample is the subset you actually measure; the population is the whole group you want to describe. Per AP Stats 3.2.B, a sample is only generalizable to the population it was selected from, so a sample of one school's students tells you about that school, not all teenagers.
Parameter (Units 5-9)
Parameters like μ, σ, and p are fixed numbers that describe the population. You almost never know them, which is exactly why inference exists. Every confidence interval and hypothesis test is an attempt to pin down a population parameter using sample data.
10% Rule (Units 5, 6, 8)
When you sample without replacement, observations aren't truly independent because the population shrinks as you go. The fix is checking that n ≤ 10% of the population size N. The CED requires this check for sampling distributions (5.7.A) and chi-square inference (8.2.E).
Sampling Distributions for Sample Means (Unit 5)
The sampling distribution of x̄ inherits its center from the population (μx̄ = μ) and its spread from the population's σ divided by √n. The population's shape even decides whether you can call x̄ approximately normal, so you can't do Topic 5.7 without knowing the population's μ, σ, and shape.
Multiple-choice questions love testing whether you know what depends on the population. Expect stems like a quality engineer measuring cable strength where the population is normal with μ = 5000 and σ = 200 (then asking about the sampling distribution of x̄), or a voter sample of 150 from a population of 1200 where you must apply the finite population correction or the 10% condition. On FRQs, the population shows up in two graded places. First, in study design questions like 2018 FRQ Q2, where a teacher samples students at her own school, you have to recognize which population the results generalize to. Second, in inference conclusions like 2017 FRQ Q5 (chi-square with a sample of 207 schizophrenia patients), where your final sentence must state a claim about the population in context, not just say 'reject H₀.' Scope of inference errors, like generalizing beyond the sampled population, cost real points.
The population is the entire group of interest; the sample is the smaller subset you actually collect data from. The mix-up that costs points is describing them with the wrong vocabulary. Numbers that describe the population are parameters (μ, σ, p) and numbers from the sample are statistics (x̄, s, p̂). If an FRQ says a researcher 'sampled 40 students from a high school with 500 students,' the 500 students are the population, the 40 are the sample, and any conclusion applies only to that school of 500.
A population is all items or subjects of interest in a study, and a sample is the subset of the population you actually measure.
Parameters (μ, σ, p) describe the population and are usually unknown; statistics (x̄, s, p̂) describe the sample and are used to estimate them.
You can only generalize results to the population the sample was actually drawn from, and only if the sample was random or otherwise representative.
The sampling distribution of x̄ has mean μ and standard deviation σ/√n, both of which come straight from the population.
If the population distribution is normal, the sampling distribution of x̄ is normal for any sample size; if not, you need a sample size of at least 30.
When sampling without replacement, check that the sample size is at most 10% of the population size so the independence condition holds.
A population is the complete set of all items or subjects a study is interested in, like every student at a school or every voter in a town. You take a sample from it and use statistics from that sample to estimate the population's parameters.
The population is everyone you want to know about; the sample is the smaller group you actually collect data from. In a question like 'a researcher samples 40 students from a high school with 500 students,' the 500 are the population and the 40 are the sample.
No. A population is any complete set of items of interest, so it can be steel cables coming off a production line, light bulbs, plastic bottles, or trees in a forest. The CED defines it as all items or subjects of interest.
Yes for generalization, no for causation. If the sample was randomly selected from the population, you can generalize findings to that population, but per AP Stats 3.2.B you cannot determine causal relationships from observational data. Only randomized experiments support cause-and-effect conclusions.
When you sample without replacement, the true standard deviation of x̄ is slightly smaller than σ/√n. The CED says that if your sample is less than 10% of the population, the difference is negligible, so checking n ≤ 0.10N lets you use the standard formula and treat observations as independent.