Two-Sample and Paired Hypothesis Testing

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/14

flashcard set

Earn XP

Description and Tags

Vocabulary flashcards covering paired samples, independent two-sample t-tests, degrees of freedom rules, F-tests for equality of variance, and related R functions.

15 Terms

1
New cards

Dependent Sample (Paired Sample)

A sample structure where each measurement in one sample has a direct one-to-one match to a measurement in another sample, such as AM and PM temperatures taken from the same individual or traits measured across twin siblings.

2
New cards

Independent Samples

Samples selected from populations that are not correlated with each other, where observations do not have a one-to-one match and sample sizes between groups are permitted to be different.

3
New cards

One-Sample t-Score

The test statistic calculated as the sample statistic minus the population mean divided by the standard error: t=xˉ−μsnt = \frac{\bar{x} - \mu}{\frac{s}{\sqrt{n}}}.

4
New cards

Central Limit Theorem Assumption for Two-Sample Testing

The requirement that both sampled populations must follow a normal distribution, or their sample sizes must each be greater than 3030 (n>30n > 30), ensuring the sampling distribution of the mean difference follows a normal distribution.

5
New cards

Null Hypothesis (H0H_0) in Two-Sample Testing

The baseline claim assuming no difference between two population means, formulated with an equal sign as μ1−μ2=0\mu_1 - \mu_2 = 0 (or involving ≥0\ge 0 or ≤0\le 0).

6
New cards

Degrees of Freedom for Paired t-Test

The degrees of freedom calculated as n−1n - 1, where nn is the matching number of paired observations across both samples.

7
New cards

Degrees of Freedom for Two-Sample t-Test (Equal Variance)

The degrees of freedom used when population variances are equal, calculated by summing the degrees of freedom from each group: n1+n2−2n_1 + n_2 - 2.

8
New cards

Degrees of Freedom for Unequal Variance (Manual Rule)

A simplified rule for manual calculations where the degree of freedom is taken as the lower value between n1−1n_1 - 1 and n2−1n_2 - 1.

9
New cards

Pooled Standard Deviation (σ^\hat{\sigma})

An estimate of standard deviation combined from both samples, utilized to calculate the standard error term in a two-sample t-test when the population variances are assumed equal.

10
New cards

Standard Error for Unequal Variance Two-Sample t-Test

The standard error term calculated without pooling variances, given by the expression: s12n1+s22n2\sqrt{\frac{s_1^2}{n_1} + \frac{s_2^2}{n_2}}.

11
New cards

F-Test

A continuous hypothesis test used to determine whether the variances of two populations are equal (σ12=σ22\sigma_1^2 = \sigma_2^2) before conducting a two-sample t-test.

12
New cards

F-Statistic

The test statistic computed as the ratio of sample variances between two groups: F=s12s22F = \frac{s_1^2}{s_2^2}.

13
New cards

F-Distribution

A family of continuous, right-skewed (non-symmetric) curves defined by the degrees of freedom of the numerator (n1−1n_1 - 1) and the denominator (n2−1n_2 - 1).

14
New cards

var.test

The function in R used to perform an F-test by taking two data vectors, outputting the F-statistic, numerator and denominator degrees of freedom, p-value, and confidence interval for the ratio of variances.

15
New cards

var.equal Parameter

An argument in R's t.test function that defaults to FALSE but must be explicitly set to TRUE if the preliminary F-test indicates that the population variances are equal.