Introduction: The Statistical "Fact-Check"

Have you ever heard a claim and thought, "That doesn't sound right"? Maybe a company claims 90% of customers are satisfied, but your small survey found much lower results. In AP Statistics, we don't just guess—we use a significance test (also called a hypothesis test) to see if a claim about a population proportion is backed up by evidence. Think of this as a formal "fact-checking" process using the power of math!

Step 1: Setting Up the Hypotheses

Before we look at the data, we have to state what we are testing. We always have two competing statements:

1. The Null Hypothesis (\(H_0\)): This is the "nothing to see here" claim. It assumes the status quo or the claimed value is true. We always write it with an equal sign.
\(H_0: p = p_0\) (where \(p_0\) is the claimed numerical value).

2. The Alternative Hypothesis (\(H_a\)): This is the claim we are actually looking for evidence to support. It’s what we suspect might actually be happening. It uses inequality symbols:
\(H_a: p > p_0\) (One-sided: we think the proportion is higher)
\(H_a: p < p_0\) (One-sided: we think the proportion is lower)
\(H_a: p \neq p_0\) (Two-sided: we think the proportion is simply different)

Important Tip: Always use the population parameter \(p\) in your hypotheses, never the sample statistic \(\hat{p}\). We are making a claim about the whole population, not just the people we talked to!

Key Takeaway: The null hypothesis is the "assumed truth" we are trying to find evidence against.

Step 2: Checking the Conditions

Don't worry if this seems like a chore—it’s the most important part of "Justifying" your results on the AP Exam! Before we calculate anything, we must verify that our model is valid.

  • Random: The data must come from a random sample or a randomized experiment. This allows us to generalize our results to the population.
  • 10% Condition: If we are sampling without replacement, our sample size \(n\) must be less than 10% of the total population (\(n \le 0.10N\)). This ensures the observations are independent enough.
  • Large Counts (Normality): We need to ensure the sampling distribution of \(\hat{p}\) is approximately Normal. Check if:
    \(n \cdot p_0 \ge 10\) and \(n \cdot (1 - p_0) \ge 10\).
    Note: Use the null value \(p_0\) here, not the sample proportion!

Step 3: Calculating the Test Statistic (\(z\))

The test statistic tells us how many standard deviations our observed sample proportion (\(\hat{p}\)) is away from the null value (\(p_0\)).

The general formula for any test statistic is:
\(\text{test statistic} = \frac{\text{statistic} - \text{parameter}}{\text{standard error}}\)

For a population proportion, we use the z-test statistic:
\(z = \frac{\hat{p} - p_0}{\sqrt{\frac{p_0(1-p_0)}{n}}}\)

Why use \(p_0\) in the denominator? Since we assume the null hypothesis is true until proven otherwise, we use that assumed proportion to calculate the standard deviation of our sampling distribution.

Did you know? A larger \(z\) score (like 3.5) means our sample result is very far from what was expected, which is strong evidence against the null hypothesis!

Step 4: Finding the p-value

The p-value is the probability of getting a sample result as extreme as, or more extreme than, the one we observed, assuming the null hypothesis is true.

  • If \(H_a\) is \(>\), the p-value is the area to the right of \(z\).
  • If \(H_a\) is \(<\), the p-value is the area to the left of \(z\).
  • If \(H_a\) is \(\neq\), the p-value is double the area in the tail.

(For more details on interpreting p-values and errors like Type I and Type II, see the dedicated chapter in this unit.)

Step 5: Drawing a Conclusion

We compare our p-value to a significance level (usually \(\alpha = 0.05\)).

If p-value \(\le \alpha\): We reject \(H_0\). We have "convincing evidence" for the alternative hypothesis \(H_a\).
If p-value \(> \alpha\): We fail to reject \(H_0\). We do not have convincing evidence for \(H_a\).

Common Mistake to Avoid: Never say you "accept the null hypothesis" or that the null is "true." We simply say we don't have enough evidence to throw it out. It's like a "Not Guilty" verdict in court—it doesn't mean the person is innocent; it just means there wasn't enough evidence to prove they were guilty.

Summary Checklist for the AP Exam

When "Carrying out a test" (Practice 3.E and 4.F), ensure you include these four components:

  1. Hypotheses: State \(H_0\) and \(H_a\) with parameters defined in context.
  2. Conditions: Check and justify the Random, 10%, and Large Counts conditions.
  3. Calculations: Identify the test (One-sample z-test for \(p\)), show the test statistic (\(z\)), and the p-value.
  4. Conclusion: Link your p-value to \(\alpha\), make a decision (reject or fail to reject), and state the conclusion in the context of the problem.

Key Takeaway: Significance testing is about measuring how "surprising" our sample data is. If it's too surprising to have happened by random chance (low p-value), we conclude the original claim was likely wrong!