Skip to main content

AP® Statistics Unit 3 flashcardsInference for Categorical Data: Proportions

34 cards · about 9 minutes for the whole deck

Flashcard drill

Space flips the card, arrow keys move, and keys 1 to 4 rate how well you knew it.

Position
1 / 34
Due
34
Mastered
0 / 34
Saved on this device · Sign in to sync

This card: New

Something wrong with this card?

What's wrong?

Please don't include personal details.

Flip the card before you rate it.

  • Estimator

    A statistic used to estimate a parameter, like p̂ for p. The value it gives from one sample is called a point estimate.

    Topic 3.1: Estimators

  • Unbiased estimator

    A statistic whose sampling distribution is centered at the parameter, so on average it neither overestimates nor underestimates.

    Topic 3.1: Estimators

  • Sampling distribution of p̂

    Mean p and standard deviation p(1−p)n\sqrt{\frac{p(1-p)}{n}}. It's approximately normal when np ≥ 10 and n(1 − p) ≥ 10.

    Topic 3.2: Sampling Distributions for Sample Proportions

  • 10% condition

    When you sample without replacement, keep n ≤ 10% of the population so the observations are close enough to independent.

    Topic 3.2: Sampling Distributions for Sample Proportions

  • Large counts condition

    The normality check for proportions: at least 10 successes and 10 failures (expected counts for a test, observed counts for an interval).

    Topic 3.2: Sampling Distributions for Sample Proportions

  • Confidence interval

    A range of plausible values for a parameter: point estimate ± margin of error.

    Topic 3.3: Constructing a Confidence Interval for a Population Proportion

  • One-sample z-interval for p

    p^±z∗⋅p^(1−p^)n\hat{p} \pm z^* \cdot \sqrt{\frac{\hat{p}(1-\hat{p})}{n}}. Use it to estimate a population proportion from one random sample.

    Topic 3.3: Constructing a Confidence Interval for a Population Proportion

  • Critical value z*

    The number of standard errors to go out on each side for a given confidence level: 1.645 for 90%, 1.960 for 95%, 2.576 for 99%.

    Topic 3.3: Constructing a Confidence Interval for a Population Proportion

  • Standard error (SE)

    An estimate of a statistic's standard deviation that uses sample values. For p̂, it's p^(1−p^)n\sqrt{\frac{\hat{p}(1-\hat{p})}{n}}.

    Topic 3.3: Constructing a Confidence Interval for a Population Proportion

  • Margin of error

    Critical value × standard error. It's how far the interval reaches on each side of the point estimate.

    Topic 3.3: Constructing a Confidence Interval for a Population Proportion

  • Sample size for a margin of error

    Solve z∗⋅p^(1−p^)n≤MEz^* \cdot \sqrt{\frac{\hat{p}(1-\hat{p})}{n}} \le \text{ME} for n, using p̂ = 0.5 if you have no estimate, and always round up.

    Topic 3.3: Constructing a Confidence Interval for a Population Proportion

  • Interpreting a confidence interval

    "We are C% confident that the interval from __ to __ captures the true [parameter in context]."

    Topic 3.4: Justifying a Claim Based on a Confidence Interval for a Population Proportion

  • Confidence level

    If you took many random samples and built an interval from each, about C% of those intervals would capture the true parameter.

    Topic 3.4: Justifying a Claim Based on a Confidence Interval for a Population Proportion

  • Interval width

    A higher confidence level makes the interval wider. A larger sample makes it narrower, and quadrupling n cuts the width in half.

    Topic 3.4: Justifying a Claim Based on a Confidence Interval for a Population Proportion

  • Null hypothesis (H₀)

    The claim of no change or no difference that a test starts out assuming, like H₀: p = 0.5. It always uses "=".

    Topic 3.5: Setting Up a Test for a Population Proportion

  • Alternative hypothesis (Hₐ)

    The claim you're looking for evidence for: p < p₀, p > p₀ or p ≠ p₀. Choose it before you see the data.

    Topic 3.5: Setting Up a Test for a Population Proportion

  • p-value

    Assuming H₀ is true, the probability of getting a statistic at least as extreme as the one observed, in the direction of Hₐ.

    Topic 3.6: p-Values

  • Significance level (α)

    The cutoff for how small a p-value must be to reject H₀, often 0.05. It's also the probability of a Type I error.

    Topic 3.7: Carrying Out a Test for a Population Proportion

  • Reject vs. fail to reject H₀

    If p-value ≤ α, reject H₀ and say there's convincing evidence for Hₐ. If not, fail to reject. You never "accept" or prove H₀.

    Topic 3.7: Carrying Out a Test for a Population Proportion

  • Test statistic for one proportion

    z=p^−p0p0(1−p0)nz = \dfrac{\hat{p} - p_0}{\sqrt{\frac{p_0(1-p_0)}{n}}}. It counts how many standard deviations p̂ is from the null value.

    Topic 3.7: Carrying Out a Test for a Population Proportion

  • Type I error

    Rejecting H₀ when it's actually true, a false alarm. Its probability is α.

    Topic 3.8: Potential Errors When Performing Tests

  • Type II error

    Failing to reject H₀ when it's actually false, a missed detection. Its probability is 1 − power.

    Topic 3.8: Potential Errors When Performing Tests

  • Power

    The probability of correctly rejecting a false H₀. It goes up with a larger n, a larger α, or a true value farther from the null value.

    Topic 3.8: Potential Errors When Performing Tests

  • Sampling distribution of p̂₁ − p̂₂

    Mean p₁ − p₂ and standard deviation p1(1−p1)n1+p2(1−p2)n2\sqrt{\frac{p_1(1-p_1)}{n_1} + \frac{p_2(1-p_2)}{n_2}}, for independent random samples or a randomized experiment.

    Topic 3.9: Sampling Distributions for the Difference Between Sample Proportions

  • Two-sample z-interval

    (p^1−p^2)±z∗⋅p^1(1−p^1)n1+p^2(1−p^2)n2(\hat{p}_1 - \hat{p}_2) \pm z^* \cdot \sqrt{\frac{\hat{p}_1(1-\hat{p}_1)}{n_1} + \frac{\hat{p}_2(1-\hat{p}_2)}{n_2}}. It estimates the difference p₁ − p₂.

    Topic 3.10: Constructing a Confidence Interval for the Difference Between Two Population Proportions

  • 0 in a difference interval

    If an interval for p₁ − p₂ contains 0, "no difference" is plausible. If it's all above or all below 0, the proportions likely differ.

    Topic 3.11: Justifying a Claim Based on a Confidence Interval for the Difference Between Two Population Proportions

  • Combined (pooled) proportion p̂c

    Total successes ÷ total sample size from both groups. A two-proportion test uses it, since H₀ says p₁ = p₂.

    Topic 3.12: Setting Up a Test for the Difference Between Two Population Proportions

  • Two-sample z-test statistic

    z=p^1−p^2p^c(1−p^c)(1n1+1n2)z = \dfrac{\hat{p}_1 - \hat{p}_2}{\sqrt{\hat{p}_c(1-\hat{p}_c)\left(\frac{1}{n_1} + \frac{1}{n_2}\right)}}. It uses the pooled proportion in the standard error.

    Topic 3.13: Carrying Out a Test for the Difference Between Two Population Proportions

  • Chi-square test for homogeneity

    Compares the distribution of one categorical variable across two or more populations or treatments, from separate samples or groups.

    Topic 3.14: Setting Up a Chi-Square Test for Homogeneity or Independence

  • Chi-square test for independence

    Checks for an association between two categorical variables measured on one sample from one population.

    Topic 3.14: Setting Up a Chi-Square Test for Homogeneity or Independence

  • Chi-square distribution

    Takes only positive values and is skewed right. Its shape depends on df, and it gets less skewed as df increases.

    Topic 3.14: Setting Up a Chi-Square Test for Homogeneity or Independence

  • Expected count

    row total×column totaltable total\dfrac{\text{row total} \times \text{column total}}{\text{table total}}: what you'd expect in a cell if H₀ were true. Every expected count should be greater than 5.

    Topic 3.15: Carrying Out a Chi-Square Test for Homogeneity or Independence

  • Chi-square statistic (χ²)

    χ2=∑(observed−expected)2expected\chi^2 = \sum \dfrac{(\text{observed} - \text{expected})^2}{\text{expected}} over every cell. Bigger values mean the data are farther from what H₀ predicts.

    Topic 3.15: Carrying Out a Chi-Square Test for Homogeneity or Independence

  • Degrees of freedom for χ²

    For a two-way table, df = (number of rows − 1)(number of columns − 1).

    Topic 3.15: Carrying Out a Chi-Square Test for Homogeneity or Independence