t-test (ttest_ind, ttest_rel) in SciPy - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the purpose of a t-test in statistics?

A t-test helps us compare the means of two groups to see if they are different in a meaningful way, considering the variation in data.

beginner

What is the difference between ttest_ind and ttest_rel in scipy?

ttest_ind compares means of two independent groups (like two different classes). ttest_rel compares means of two related groups (like before and after measurements on the same people).

beginner

In a t-test, what does the p-value tell us?

The p-value shows the chance that the difference we see happened by random luck. A small p-value (usually less than 0.05) means the difference is likely real.

beginner

When should you use a paired t-test (ttest_rel)?

Use a paired t-test when the two sets of data are linked or matched, like measuring the same people before and after a treatment.

intermediate

What assumptions do t-tests make about the data?

T-tests assume the data is roughly normally distributed and that the groups have similar variances. For ttest_ind, the groups should be independent.

Which scipy function compares means of two independent samples?

A chi2_contingency

B ttest_rel

C anova

D ttest_ind

ttest_ind is used for independent samples.

When should you use ttest_rel?

A Comparing more than two groups

B Comparing two unrelated groups

C Comparing two related or paired groups

D Testing categorical data

ttest_rel is for related or paired samples.

What does a p-value less than 0.05 usually mean in a t-test?

A No difference between groups

B Strong evidence of a difference

C Data is not normal

D Test failed

A p-value below 0.05 suggests the difference is statistically significant.

Which assumption is NOT required for a t-test?

A Data must be categorical

B Groups have equal variances

C Data is normally distributed

D Samples are independent for <code>ttest_ind</code>

T-tests require numerical data, not categorical.

If you measure the same group twice, which test is best?

A ttest_rel

B correlation test

C chi-square test

D ttest_ind

Repeated measures on the same group use ttest_rel.

Explain the difference between ttest_ind and ttest_rel and give an example of when to use each.

Think about whether the data points in the two groups are connected or not.

Describe what a p-value means in the context of a t-test and how you decide if the difference is significant.

Consider how confident you are that the groups truly differ.