t-test with scipy.stats in Data Analysis Python - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is a t-test used for in data analysis?

A t-test helps us compare the average values of two groups to see if they are different in a meaningful way.

beginner

Which function in scipy.stats is commonly used to perform a t-test for independent samples?

The function scipy.stats.ttest_ind() is used to compare the means of two independent groups.

beginner

What does the p-value from a t-test tell us?

The p-value shows the chance that the difference between groups happened by random chance. A small p-value (like less than 0.05) means the difference is likely real.

intermediate

How do you perform a paired t-test using scipy.stats?

Use scipy.stats.ttest_rel() to compare two related samples, like measurements before and after a treatment on the same subjects.

intermediate

What assumptions must be met to correctly use a t-test?

The data should be roughly normally distributed, the samples should be independent (except for paired tests), and variances should be similar for independent t-tests.

Which scipy.stats function is used for an independent two-sample t-test?

A ttest_1samp

B ttest_rel

C anova

D ttest_ind

ttest_ind compares means of two independent groups. ttest_rel is for paired samples, ttest_1samp compares one sample to a known value, and ANOVA is a different test.

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

A There is strong evidence the groups differ

B The groups are definitely the same

C The test was done incorrectly

D The sample size is too small

A p-value below 0.05 suggests the difference between groups is statistically significant, meaning it is unlikely due to chance.

Which t-test should you use to compare measurements before and after treatment on the same people?

A ttest_ind

B ttest_rel

C ttest_1samp

D chi-square test

ttest_rel is for paired samples, like before-and-after measurements on the same subjects.

Which assumption is NOT required for a t-test?

A Data is normally distributed

B Samples are independent (for independent t-test)

C Data is categorical

D Variances are similar between groups

T-tests require numerical data, not categorical data.

What does scipy.stats.ttest_1samp() do?

A Tests if a sample mean differs from a known value

B Compares two related groups

C Compares two independent groups

D Tests for correlation between variables

ttest_1samp tests if the average of one sample is different from a specific number.

Explain how to perform an independent t-test using scipy.stats and interpret the results.

Think about comparing two groups like test scores from two classes.

Describe the main assumptions behind using a t-test and why they matter.

Consider what could go wrong if data is very skewed or samples are related but treated as independent.