Confidence Intervals: Difference of Two Proportions - AP Statistics

1. Explain why independence is important for constructing confidence intervals for two proportions.

   Independence ensures samples do not affect each other. Prevents bias from one sample influencing the other.

2. Identify the correct formula for pooled standard error for two proportions.

   $\sqrt{\hat{p}(1-\hat{p})(\frac{1}{n_1} + \frac{1}{n_2})}$. Uses pooled proportion for hypothesis testing standard error.

3. Find the value of $x_2$ for $p_2 = 0.3$ and $n_2 = 200$.

   $x_2 = 0.3 \times 200 = 60$. Number of successes equals proportion times sample size.

4. Predict the confidence interval if $z^*$ increases and sample sizes remain constant.

   The confidence interval widens. Larger critical value increases margin of error.

5. What does $p_1$ represent in the confidence interval formula for two proportions?

   $p_1$ is the sample proportion of the first group. The proportion of successes in group 1.

6. State the null hypothesis for a test comparing two proportions.

   $H_0: p_1 = p_2$. Assumes no difference between the two population proportions.

7. Identify the conditions required for using the confidence interval for two proportions.

   Random sampling, normality, and independent samples. Ensures valid normal approximation and unbiased estimates.

8. Given $p_1 = 0.3$, $p_2 = 0.5$, $n_1 = 100$, $n_2 = 120$, calculate the standard error.

   $\sqrt{\frac{0.3 \times 0.7}{100} + \frac{0.5 \times 0.5}{120}}$. Standard error combines variability from both samples.

9. Calculate the confidence interval width for $p_1 = 0.7$, $p_2 = 0.5$, $n_1 = 80$, $n_2 = 90$.

   $2 \times z^* \times \text{SE}$. Width equals twice the margin of error.

10. Calculate the lower bound of the interval for $p_1 = 0.8$, $p_2 = 0.6$, $n_1 = 100$, $n_2 = 120$.

    $(0.8 - 0.6) - z^* \times \text{SE}$. Lower bound subtracts margin of error from difference.

11. What is the formula for the confidence interval of the difference between two proportions?

    $(p_1 - p_2) \, \pm \, z^* \cdot \sqrt{\frac{p_1(1-p_1)}{n_1} + \frac{p_2(1-p_2)}{n_2}}$. Estimates difference $(p_1 - p_2)$ with margin of error using normal distribution.

12. What happens to the confidence interval if the sample proportions are closer to 0.5?

    The interval becomes wider. Proportions near 0.5 have maximum variance.

13. Determine the sample proportions $p_1$ and $p_2$ given successes and sample sizes.

    $p_1 = \frac{x_1}{n_1}$, $p_2 = \frac{x_2}{n_2}$. Sample proportion equals successes divided by sample size.

14. What does it mean if a confidence interval for two proportions includes zero?

    There may be no significant difference between the proportions. Zero in interval suggests proportions could be equal.

15. Identify the margin of error in the confidence interval formula for two proportions.

    $z^* \cdot \sqrt{\frac{p_1(1-p_1)}{n_1} + \frac{p_2(1-p_2)}{n_2}}$. The maximum likely error in the estimate of $p_1 - p_2$.

16. Calculate the standard error with $p_1 = 0.4$, $n_1 = 150$, $p_2 = 0.3$, $n_2 = 200$.

    $\sqrt{\frac{0.4 \times 0.6}{150} + \frac{0.3 \times 0.7}{200}}$. Standard error formula with given proportions and sample sizes.

17. What is the effect of overlapping confidence intervals on the difference between proportions?

    Overlapping intervals suggest no significant difference. Overlapping intervals indicate similar population proportions likely.

18. Calculate the upper bound of the interval for $p_1 = 0.7$, $p_2 = 0.5$, $n_1 = 150$, $n_2 = 130$.

    $(0.7 - 0.5) + z^* \times \text{SE}$. Upper bound adds margin of error to difference.

19. What is the meaning of $p_2$ in the confidence interval formula for two proportions?

    $p_2$ is the sample proportion of the second group. The proportion of successes in group 2.

20. What is a Type I error in the context of comparing two proportions?

    Rejecting $H_0$ when $H_0$ is true. False positive: concluding difference when none exists.

21. What is a Type II error in the context of comparing two proportions?

    Failing to reject $H_0$ when $H_a$ is true. False negative: missing a real difference between proportions.

22. What is the pooled proportion in a significance test for two proportions?

    $\hat{p} = \frac{x_1 + x_2}{n_1 + n_2}$. Combines both samples assuming equal population proportions.

23. Identify the error: Using $t^*$ instead of $z^*$ for two proportions confidence interval.

    Correct use: $z^*$ for proportions. Use $z^*$ for proportions, not $t^*$ which is for means.

24. State the alternative hypothesis for a test comparing two proportions.

    $H_a: p_1 \neq p_2$. Two-sided test for any difference between proportions.

25. Define $n_2$ in the context of confidence intervals for two proportions.

    $n_2$ is the sample size of the second group. Number of observations in group 2.

26. How does increasing the confidence level affect the width of the confidence interval?

    Increasing the confidence level widens the confidence interval. Higher confidence requires larger critical value $z^*$.

27. What is the effect of a larger sample size on the width of the confidence interval?

    Larger sample sizes result in narrower confidence intervals. More data reduces sampling variability and uncertainty.

28. Find the value of $x_1$ for $p_1 = 0.4$ and $n_1 = 150$.

    $x_1 = 0.4 \times 150 = 60$. Number of successes equals proportion times sample size.

29. Find $p_1 - p_2$ for $p_1 = 0.6$ and $p_2 = 0.4$.

    $p_1 - p_2 = 0.2$. Simple subtraction: $0.6 - 0.4 = 0.2$.

30. Calculate the critical value $z^*$ for a 95% confidence interval.

    $z^* \approx 1.96$. Standard critical value for 95% confidence level.

31. Define $n_1$ in the context of confidence intervals for two proportions.

    $n_1$ is the sample size of the first group. Number of observations in group 1.

32. What is the role of a critical value in a confidence interval?

    It determines the margin of error for the interval. Critical value multiplied by standard error gives margin of error.

33. Estimate the 95% confidence interval for $p_1 = 0.5$, $p_2 = 0.4$, $n_1 = 100$, $n_2 = 100$.

    $(0.5 - 0.4) \pm 1.96 \times \text{SE}$. Difference plus/minus critical value times standard error.

34. What does $z^*$ represent in the confidence interval formula for two proportions?

    $z^*$ is the critical value from the standard normal distribution. Corresponds to the desired confidence level.

35. What is the typical confidence level used in practice for a confidence interval?

    Typical confidence levels are 90%, 95%, and 99%. Balance between confidence and precision in estimation.

36. What is the purpose of constructing a confidence interval for the difference of two proportions?

    To estimate the difference between two population proportions. Provides a range of plausible values for $p_1 - p_2$.

Study these flashcards