Topic 5.1 Introducing Statistics: Why Is My Sample Not Like Yours? Recognize sampling variability, the difference between a parameter and a statistic, and that a statistic varies from sample to sample in a predictable way.

What this topic is asking

The College Board (Topic 5.1) wants you to recognize sampling variability, distinguish a parameter from a statistic, and understand that a statistic varies from sample to sample in a predictable way, the idea that motivates the whole unit.

Parameter versus statistic

The notation encodes the distinction: Greek-style or unadorned symbols ($\mu$, $p$, $\sigma$) for population parameters, and English letters or hatted symbols ($\bar{x}$, $\hat{p}$, $s$) for sample statistics. The whole point of inference is that the parameter is fixed but unknown, while the statistic is known but variable, so we use the variable statistic to estimate the fixed parameter, accounting for the variability.

Sampling variability

This reframes "why is my sample not like yours?" The answer is that two honest random samples should differ, because they are different subsets of the population. The achievement of the unit is to show that this variability is lawful: a statistic does not bounce around arbitrarily but follows a describable distribution, so we can say how close to the parameter a typical sample statistic falls.

Why predictable variability is the key

The reason Topic 5.1 opens the unit is that predictable variability is what makes inference possible at all. If a sample statistic varied wildly and unpredictably, a single sample would tell us nothing reliable about the population. But because the sampling distribution has a known center (typically the parameter itself, for an unbiased statistic), a known spread (which shrinks as $n$ grows), and often a known shape (approximately normal, by the central limit theorem), we can quantify how far a sample statistic is likely to be from the truth. That quantification is exactly what a confidence interval and a significance test do. So the apparently humble observation that samples differ is the seed of everything that follows: every interval and every test in Units 6 through 9 is, at heart, a statement about where a statistic falls in its sampling distribution. Understanding that a statistic is itself a random variable with a distribution, rather than a single fixed number, is the conceptual leap Topic 5.1 asks you to make.

Connecting to what came before

Two earlier threads converge here. From Unit 3, random sampling is what makes the sampling distribution well-behaved: only when samples are drawn randomly is the statistic's variation governed by known probability, so the whole unit assumes random selection. From Unit 4, a statistic computed from a random sample is a random variable (its value depends on the chance of which sample is drawn), so the mean, standard deviation, and shape ideas of Topic 4.7 to 4.9 apply directly to it. The sampling distribution of $\bar{x}$ or $\hat{p}$ is just the probability distribution of that random variable. Seeing Unit 5 as "apply the random-variable machinery of Unit 4 to a statistic computed from a random sample of Unit 3" ties the course together and explains why the formulas in the coming topics, the mean and standard deviation of $\hat{p}$ and $\bar{x}$, look like the combining rules you already know.

Try this

Q1. Distinguish a parameter from a statistic, with an example of each. [2 points]

• Cue. A parameter describes a population and is fixed (for example the true proportion $p$); a statistic is computed from a sample and varies (for example the sample proportion $\hat{p}$).

Q2. Two random samples from the same population give different proportions. Is this a problem? Explain. [1 point]

• Cue. No; it is sampling variability, the expected result of different samples containing different individuals, and it follows a predictable sampling distribution.