AP® Statistics Unit 4 flashcardsInference for Quantitative Data: Means
27 cards · about 7 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 / 27
- Due
- 27
- Mastered
- 0 / 27
This card: New
Flip the card before you rate it.
Sampling distribution of x̄
For random samples of size n: mean μ and standard deviation . Bigger samples give sample means that vary less.
Topic 4.1: Sampling Distributions for Sample Means
Normal shape for x̄
x̄ is normal if the population is normal, and approximately normal if n ≥ 30 (by the CLT). A strongly skewed population may need a bigger n.
Topic 4.1: Sampling Distributions for Sample Means
t-distribution
Symmetric and bell-shaped like the standard normal, but with heavier tails. As df increases, it gets closer to the standard normal.
Topic 4.2: Constructing a Confidence Interval for a Population Mean or Population Mean Difference
Degrees of freedom (one sample)
df = n − 1 for one-sample and paired t procedures. It tells you which t-distribution to use.
Topic 4.2: Constructing a Confidence Interval for a Population Mean or Population Mean Difference
Critical value t*
The t-value that matches your confidence level and df. It's bigger than z* for the same confidence, especially when df is small.
Topic 4.2: Constructing a Confidence Interval for a Population Mean or Population Mean Difference
Standard error of x̄
. It estimates the standard deviation of x̄ when you don't know σ, which is almost always.
Topic 4.2: Constructing a Confidence Interval for a Population Mean or Population Mean Difference
One-sample t-interval
with df = n − 1. It estimates a population mean μ.
Topic 4.2: Constructing a Confidence Interval for a Population Mean or Population Mean Difference
Paired data
Two measurements on the same unit or on matched units, like before and after scores. Analyze the differences as one sample.
Topic 4.2: Constructing a Confidence Interval for a Population Mean or Population Mean Difference
Mean difference (μd)
The population mean of the paired differences. A paired t-interval or t-test works on the sample of differences, using and .
Topic 4.2: Constructing a Confidence Interval for a Population Mean or Population Mean Difference
Interpreting a t-interval
"We are C% confident that the interval from __ to __ captures the true mean [variable in context]." It's about μ, not about individual values.
Topic 4.3: Justifying a Claim Based on a Confidence Interval for a Population Mean or Population Mean Difference
Hypotheses for a mean
H₀: μ = μ₀, with Hₐ: μ < μ₀, μ > μ₀ or μ ≠ μ₀. For paired data, H₀: .
Topic 4.4: Setting Up a Test for a Population Mean or Population Mean Difference
Conditions for t procedures
Random sample or randomized experiment; 10% condition when sampling without replacement; and a normal population, n ≥ 30, or no strong skew or outliers in the sample.
Topic 4.4: Setting Up a Test for a Population Mean or Population Mean Difference
One-sample t statistic
, with df = n − 1. It counts how many standard errors x̄ is from the null value.
Topic 4.5: Carrying Out a Test for a Population Mean or Population Mean Difference
Sampling distribution of x̄₁ − x̄₂
Mean μ₁ − μ₂ and standard deviation , for independent random samples or a randomized experiment.
Topic 4.6: Sampling Distributions for the Difference Between Two Sample Means
Variances add
For independent samples, the variance of x̄₁ − x̄₂ is the sum of the two variances, even though you're subtracting the means.
Topic 4.6: Sampling Distributions for the Difference Between Two Sample Means
Two-sample t-interval
. Technology gives the df. It estimates μ₁ − μ₂.
Topic 4.7: Constructing a Confidence Interval for the Difference Between Two Population Means
Independent vs. paired samples
Use two-sample t procedures for two separate groups. Use paired t procedures when each value in one group is linked to one value in the other.
Topic 4.9: Setting Up a Test for the Difference Between Two Population Means
Hypotheses for two means
H₀: μ₁ = μ₂ (or μ₁ − μ₂ = 0), with a one-sided or two-sided Hₐ. Define both parameters in context.
Topic 4.9: Setting Up a Test for the Difference Between Two Population Means
Two-sample t statistic
. Use technology for the df and the p-value.
Topic 4.10: Carrying Out a Test for the Difference Between Two Population Means
0 in an interval for μ₁ − μ₂
If 0 is inside the interval, "no difference" in means is plausible. If the whole interval is above or below 0, the means likely differ.
Topic 4.8: Justifying a Claim Based on a Confidence Interval for the Difference Between Two Population Means
Randomized experiment conclusion
With random assignment, a significant difference supports a cause-and-effect conclusion, but only for units like those in the study.
Topic 4.10: Carrying Out a Test for the Difference Between Two Population Means
Robustness of t procedures
t procedures still work well when the population isn't exactly normal, as long as there's no strong skew or outliers, especially for larger n.
Topic 4.4: Setting Up a Test for a Population Mean or Population Mean Difference
Standard error vs. standard deviation
The standard deviation s describes how individual values vary. The standard error describes how x̄ varies from sample to sample.
Topic 4.2: Constructing a Confidence Interval for a Population Mean or Population Mean Difference
Effect of n on a t-interval
A larger n shrinks and also shrinks t* (more df), so the interval gets narrower.
Topic 4.3: Justifying a Claim Based on a Confidence Interval for a Population Mean or Population Mean Difference
Paired t-test
A one-sample t-test on the differences, with H₀: and , df = n − 1.
Topic 4.5: Carrying Out a Test for a Population Mean or Population Mean Difference
Matched pairs t-interval
with df = n − 1, where and come from the differences. It estimates .
Topic 4.2: Constructing a Confidence Interval for a Population Mean or Population Mean Difference
Conditions for a two-sample t-interval
Two independent random samples (or random assignment), 10% condition for each sample, and both n ≥ 30 or no strong skew or outliers in either sample.
Topic 4.7: Constructing a Confidence Interval for the Difference Between Two Population Means