Lesson 3: 8.3 Estimating a Difference of Proportions

1. How do we combine the variability of two different samples?

2. What does a confidence interval containing zero tell us about the difference between the groups?

3. How do the conditions change when we have two independent groups?

AP Stats CED: VAR-6.B, UNC-4.I (Two-sample intervals for proportions). Common Core: HSS-IC.B.4, HSS-IC.B.6.

Description
This section covers the Two-Sample z-Interval for p1-p2. Students learn to calculate the standard error for the difference. They use the same z* critical values but must check the Random, 10%, and Large Counts conditions for both samples independently.

Purpose
To allow students to compare two distinct populations (e.g., men vs. women, Treatment A vs. Control) to see if there is evidence of a real-world difference.

DOK Level
Level 3 (Strategic Thinking): Students must interpret the interval in context, specifically evaluating whether the interval provides evidence of a difference (i.e., does it exclude 0?).

Struggling Learners: Focus on the "Zero Rule." Create a visual "Number Line" poster. If the interval is (0.02, 0.10), both numbers are positive, Group 1 is likely higher. If it’s (-0.05, 0.04), zero is inside we can't claim a difference.

Advanced Learners: Have them explore why we add the variances in the standard error formula even though we are subtracting the proportions. This reinforces the "Pythagorean Theorem of Statistics" VAR(X-Y) = VarX + VarY.

ELL Learners: Use a "Comparison Chart" to organize data from Sample 1 and Sample 2. Labeling "Group A" and "Group B" clearly before starting the formula helps prevent mixing up the p-hat and n values in the long square root formula.

Application activity to real-world data