A researcher draws one random sample of 50 people and calculates the sample mean. Which statement about the sampling distribution of the mean is correct?
AThe sampling distribution only exists if the researcher actually draws many samples in practice
BThe sampling distribution is the distribution of individual scores within the researcher's one sample
CThe sampling distribution describes how the sample mean would vary across all possible samples of size 50, even though only one was drawn
DThe sampling distribution cannot be defined without knowing the true population distribution
The sampling distribution is a theoretical object — it describes what would happen if you repeated the sampling process infinitely. It exists regardless of whether you draw one sample or a thousand; it is a property of the sampling procedure and the population, not of how many times you run the experiment. Option B confuses the sampling distribution with the within-sample distribution of individual scores, which is a different object entirely.
Question 2 Multiple Choice
A statistics textbook states: 'The standard error of the mean is σ/√n.' What does this quantity measure?
AThe typical distance between individual observations and the population mean μ
BThe typical error introduced when estimating σ from sample data
CThe standard deviation of the sampling distribution of the sample mean
DThe half-width of the 95% confidence interval for the mean
σ/√n is the standard deviation of the sampling distribution of X̄ — it measures how much the sample mean varies from sample to sample, not how much individual values vary within one sample. Option A describes σ (population standard deviation), not the standard error. Option D is related (the CI uses the standard error) but is not what the quantity directly measures. The distinction matters: standard error shrinks as n grows because larger samples produce more consistent estimates.
Question 3 True / False
The sampling distribution of the sample mean has smaller spread (standard deviation) when sample size n is larger, because averaging more observations reduces the influence of extreme values.
TTrue
FFalse
Answer: True
The standard error σ/√n decreases as n increases. Intuitively, a sample mean based on 1,000 observations will almost always be close to the population mean, because unusually high and low values tend to cancel out. A sample mean based on 5 observations might easily deviate far from μ by chance. This is why larger samples yield more precise estimates — they compress the sampling distribution around the true parameter.
Question 4 True / False
The sampling distribution of the sample mean describes how individual observations are distributed within a single sample.
TTrue
FFalse
Answer: False
The sampling distribution is about the statistic (the mean), not about individual data points. Within-sample variability is captured by the sample's standard deviation. The sampling distribution answers a different question: if I drew many samples of the same size and computed the mean each time, what distribution would those means follow? These are two distinct distributions — confusing them is one of the most common errors in understanding statistical inference.
Question 5 Short Answer
Why is the sampling distribution described as a 'theoretical object,' and why does this matter for the logic of statistical inference?
Think about your answer, then reveal below.
Model answer: In practice, we almost always draw only one sample. The sampling distribution is theoretical because it describes what would happen across infinitely repeated samples — a process we never actually complete. It matters because all of frequentist inference is built on it: a p-value asks 'how often would I see a statistic this extreme if the null hypothesis were true?' — which is a question about the sampling distribution under the null. Confidence intervals, hypothesis tests, and standard errors all implicitly reference this theoretical distribution. Without it, statements about 'what would happen by chance' have no foundation.
This is why sampling distributions are foundational but often misunderstood: students see one dataset and ask 'where is the sampling distribution?' It doesn't appear in your data — it is a theoretical construct that quantifies the uncertainty introduced by the random act of sampling, allowing you to make probability statements about a single observed result.