01 / 21

Carrying Out a Test
for a Population Mean

Master the one-sample \(t\)-test: from stating hypotheses to computing \(p\)-values and drawing conclusions about \(\mu\).

AP Statistics

What Is a Hypothesis Test?

A hypothesis test is a formal procedure for deciding whether sample data provide convincing evidence against a claim about a population parameter. When the parameter is a population mean \(\mu\), we use the one-sample \(t\)-test.

\(H_0\) : Null Hypothesis
The "no change" claim: the population mean equals a specific value \(\mu_0\).
\(H_a\) : Alternative Hypothesis
What we suspect is true: \(\mu \neq \mu_0\), \(\mu > \mu_0\), or \(\mu < \mu_0\).

The Big Picture of a \(t\)-Test

STATE Hypotheses PLAN Check Conditions DO Calculate \(t\), \(p\) CONCLUDE Decision + Context

Why the \(t\)-Distribution?

When the population standard deviation \(\sigma\) is unknown (which is almost always), we replace it with the sample standard deviation \(s\). This extra uncertainty makes the sampling distribution wider than a normal curve: it follows a \(t\)-distribution with \(df = n - 1\).

0 \(t\) \(z\) (normal) \(t\) (\(df = 3\)) Heavier tails = more area in the extremes

The Test Statistic Formula

$$ t = \frac{\bar{x} - \mu_0}{\dfrac{s}{\sqrt{n}}} $$

\(\bar{x}\)
Sample mean: the observed average from your data.
\(\mu_0\)
Hypothesized value of the population mean under \(H_0\).
\(s\)
Sample standard deviation: measures spread in the sample.
\(n\)
Sample size: number of observations collected.

Conditions for the \(t\)-Test

1
Random: The data come from a random sample or a randomized experiment.
2
Independence (10% Rule): When sampling without replacement, the sample size \(n\) must be less than 10% of the population \(N\).
3
Normal/Large Sample: The population is approximately normal, or \(n \geq 30\) (CLT), or graph of data shows no strong skewness or outliers.

One-Tailed vs Two-Tailed Tests

One-Tailed Two-Tailed
\(H_a\) \(\mu > \mu_0\) or \(\mu < \mu_0\) \(\mu \neq \mu_0\)
Tail(s) shaded 1 tail (right or left) Both tails
\(p\)-value Area in one tail \(2 \times\) area in one tail
When to use Directional claim "Different from" claim

Understanding the \(p\)-Value

The \(p\)-value is the probability of obtaining a test statistic as extreme as (or more extreme than) the observed value, assuming \(H_0\) is true.

0 \(t_{obs}\) \(p\)-value

How to Carry Out the Test

1
State hypotheses: Write \(H_0\!: \mu = \mu_0\) and choose \(H_a\) (with direction).
2
Check conditions: Random, Independent (10%), Normal/Large Sample.
3
Compute the test statistic: \(\displaystyle t = \frac{\bar{x} - \mu_0}{s / \sqrt{n}}\) with \(df = n - 1\).
4
Find the \(p\)-value: Use a \(t\)-table or calculator: tcdf with appropriate bounds.
5
Conclude: Compare \(p\) to \(\alpha\). State your decision in context of the problem.

Example 1: Coffee Shop Wait Times

A coffee shop claims the mean wait time is \(\mu = 4\) minutes. A random sample of \(n = 25\) customers has \(\bar{x} = 4.8\) min and \(s = 1.5\) min. At \(\alpha = 0.05\), is there evidence the mean wait is longer than 4 minutes?

Example 1: Finding the \(p\)-Value

0 \(t = 2.667\) \(p \approx 0.007\)

Example 2: Body Temperature

Is the mean human body temperature different from \(98.6^\circ\text{F}\)? A random sample of \(n = 50\) healthy adults has \(\bar{x} = 98.26\) and \(s = 0.73\). Test at \(\alpha = 0.05\).

Example 2: Two-Tailed \(p\)-Value

0 \(-3.29\) \(+3.29\) \(p/2\) \(p/2\)

Example 3: Failing to Reject \(H_0\)

A battery manufacturer claims batteries last \(\mu = 500\) hours. A sample of \(n = 16\) batteries gives \(\bar{x} = 492\) hours and \(s = 25\) hours. Test \(H_a\!: \mu < 500\) at \(\alpha = 0.05\).

Common Mistakes to Avoid

Using \(z\) Instead of \(t\)
When \(\sigma\) is unknown and you use \(s\), you must use the \(t\)-distribution, not \(z\).
"Accept \(H_0\)"
Never say "accept \(H_0\)." The correct phrase is "fail to reject \(H_0\)." We lack evidence, not proof.
No Context in Conclusion
Always state your conclusion in terms of the problem: reference the variable, units, and population.

Summary: Decision Rules

Scenario Decision Template Conclusion
\(p < \alpha\) Reject \(H_0\) There is convincing evidence that \(\mu\) [direction] \(\mu_0\).
\(p \geq \alpha\) Fail to reject \(H_0\) There is not convincing evidence that \(\mu\) [direction] \(\mu_0\).

Question 1 of 5

A researcher tests \(H_0\!: \mu = 50\) vs \(H_a\!: \mu > 50\) with a sample of \(n = 36\), \(\bar{x} = 53\), and \(s = 9\). What is the value of the test statistic?

Question 2 of 5

A \(t\)-test yields \(p = 0.032\) with \(\alpha = 0.05\). What is the correct conclusion?

Question 3 of 5

For a two-tailed \(t\)-test, a calculator gives a one-tail area of \(0.018\). What is the \(p\)-value?

Question 4 of 5

Which of the following is not a condition that must be verified before performing a one-sample \(t\)-test?

Question 5 of 5

A sample of \(n = 20\) gives \(t = -1.85\) for \(H_a\!: \mu < 100\) with \(df = 19\). Using a \(t\)-table, the \(p\)-value is between 0.025 and 0.05. At \(\alpha = 0.05\), what do you conclude?