Statistics Fundamentals cookbook

8 recipes · 50 steps · about 43 minutes of reading

Compare two groups' averages with a t-test
Two channels look different in average order value, and I can't tell if that's a real gap or just noise.
BeginnerRead 5 minDoing: an hour6 steps
An A/B test isn't significant, and the team still needs a call
My test result came back not significant, and my manager still wants to know what to do next.
Beginner to intermediateRead 5 minDoing: an hour7 steps
Work out how big a sample an A/B test needs
I want to launch a test tomorrow, and I have no idea how many visitors it actually needs to give a trustworthy answer.
Beginner to intermediateRead 5 minDoing: under an hour5 steps
Judge whether a spike in daily counts is real with a Poisson check
Support tickets jumped yesterday, and I need to know if that's a real problem or just an ordinary busy day.
IntermediateRead 5 minDoing: 30 minutes6 steps
State a confidence interval without overclaiming it
I told my manager the confidence interval means we're 95% sure, and now I'm not sure that's actually true.
IntermediateRead 5 minDoing: 30 minutes5 steps
Your data fails normality; try a nonparametric test instead
My data is full of outliers and skewed values, and the usual t-test rules feel like they don't apply anymore.
IntermediateRead 5 minDoing: an hour7 steps
Find which group differs with a Tukey post-hoc test
My ANOVA says the three groups aren't all equal, but it won't tell me which ones actually differ.
Intermediate to advancedRead 6 minDoing: an hour8 steps
Which of your dashboard's flagged metrics survive a correction for multiple testing?
Twelve metrics on the dashboard, and one just turned green. Is that a real win, or is that just what happens when you check twelve things at once?
Intermediate to advancedRead 7 minDoing: about an hour6 steps

Cookbook badge

Finish all 8 recipes to earn this.

In these roadmaps

Data Analyststep 10 of 14

Statistics Fundamentals cookbook

Compare two groups' averages with a t-test

An A/B test isn't significant, and the team still needs a call

Work out how big a sample an A/B test needs

Judge whether a spike in daily counts is real with a Poisson check

State a confidence interval without overclaiming it

Your data fails normality; try a nonparametric test instead

Find which group differs with a Tukey post-hoc test

Which of your dashboard's flagged metrics survive a correction for multiple testing?

In these roadmaps

Statistics Fundamentals cookbook

Compare two groups' averages with a t-test

An A/B test isn't significant, and the team still needs a call

Work out how big a sample an A/B test needs

Judge whether a spike in daily counts is real with a Poisson check

State a confidence interval without overclaiming it

Your data fails normality; try a nonparametric test instead

Find which group differs with a Tukey post-hoc test

Which of your dashboard's flagged metrics survive a correction for multiple testing?

In these roadmaps