Jun 20

6 min read

Data Science

Being Normal Is Overrated

…for t-tests, at least

Pop quiz.

Can you use a t-test to tell if these two distributions are drawn from populations with the same expected value?

More from Naim Kabir

Engineer. Focused on experimentation, causal inference, and good software design. naimkabir.com

Love podcasts or audiobooks? Learn on the go with our new app.

Try Knowable

Recommended from Medium

Mark Pfeiffer

SiaSearch (now Scale Nucleus)

Effortlessly explore the nuScenes dataset with SiaSearch

Ed Summers

Archival Attention

Nate Barlow

Analytics Vidhya

Python for playing around with the Floral Reflectance Database (FReD)

Theo McCaie

Met Office Informatics Lab

How to (and not to) handle metadata in high momentum datasets

Alif Luqman Kahfi

I AM GROWTH

Sarah Fleckenstein

Atlanta’s Coffee Priorities: Starbucks, Dunkin’, or Local?

Lilia Tovbin

When Will The Pandemic End in The US? Soon, in Most Places.

US top states deaths from Covid-19 trends — August 2021

Santhosh Kumar M

Building a trustworthy data(team) — Part 1

About Help Terms Privacy

Data Science

Being Normal Is Overrated

…for t-tests, at least

More from Naim Kabir

Recommended from Medium

Effortlessly explore the nuScenes dataset with SiaSearch

Archival Attention

Python for playing around with the Floral Reflectance Database (FReD)

How to (and not to) handle metadata in high momentum datasets

I AM GROWTH

Atlanta’s Coffee Priorities: Starbucks, Dunkin’, or Local?

When Will The Pandemic End in The US? Soon, in Most Places.

Building a trustworthy data(team) — Part 1

Get the Medium app

Naim Kabir

More from Medium

A tricky part of Applied Data Science: SELECTION BIAS

Ridge and Lasso Regression

How To Be A Great Data Tutor

Time Series Forecasting of previous years UPI datasets(2016–2021)