Compare groups carefully · AP Statistics

Comparing groups requires a clear parameter: a difference in proportions, a difference in means, or an association between categorical variables. The design determines which observations can be treated as independent.

A reliable method

For independent means, combine squared standard errors before taking a square root. For paired data, analyze each pair’s difference rather than treating the two columns as independent groups.

The common trap

For a contingency table under independence, expected count equals row total times column total divided by grand total. The chi-square contribution is squared observed-minus-expected difference divided by expected count.

Make the connection

Check procedure-specific conditions. Do not use a chi-square contribution as a stand-alone p-value, and do not turn an observational association into a causal conclusion.

Pairing changes the analysis unit

For paired observations, one difference per person is the relevant response. Independent groups require a different uncertainty calculation. If the two independent mean estimates have SE 3 and SE 4, their difference has SE √(9+16)=5. Subtracting means does not mean subtracting their variances.

Counts need a model for comparison

A table cell with row total 40, column total 30 and total 100 has expected count 12 under independence. If observed count is 18, its chi-square contribution is (18−12)²/12=3. One cell contribution is not the full test statistic, and neither number alone is a p-value. A tail probability requires the relevant sampling distribution and its degrees of freedom.

Transfer check

Name the parameter before selecting an inference procedure: two population proportions, two population means, or association among categorical variables. For each, state whether samples are independent or paired and which conditions you would check. The numerical arithmetic in this pilot does not replace that full justification.

WORKED EXAMPLE

Two independent groups have sample proportions 3/5 and 3/10. What is the estimated difference, first minus second?

Keep the direction of subtraction fixed.
Subtract 3/10 from 3/5.
The estimated difference is 3/10.

Answer: 3/10

Compare the other answer choices

9/10

Add the group proportions.

3/5

Report the first proportion without subtracting the second group.

7/10

Complement the desired difference.

Try it, then explain it.

Complete a short set without looking at the example. After each answer, explain which operation preserved the relationship and why the result makes sense.

Practice this skill

Sources & authorship

Original Slope explanations and examples. References support the concepts; they are not the source of copied questions. Independent educator review is pending.

AP Statistics 2026–27 course framework