Tweet

Cassie Kozyrkov

13 Jan, 7 tweets, 1 min read

What are the 5 most important questions to ask when starting with a new dataset? A video ( + a thread 🧵)

bit.ly/quaesita_yt5da…

#1 - Purpose

What was the purpose for which this dataset was collected?

#2 - Competence

How competent were the people who collected this data?

#3 - Agenda

Is there any reason to suspect that the agenda of the people who collected it might bias the data?

#4 - Clarity

How clear is the documentation accompanying the data? Can you be sure you know what happened in the real world when the data were captured?

#5 - Processing

Are you looking at raw data or has the information been transformed in some way? Does this render the data unsuitable for your needs?

For my quick guide to working with inherited data, take a look at this bit.ly/quaesita_notyo…

• • •

Missing some Tweet in this thread? You can try to force a refresh

This Thread may be Removed Anytime!

Twitter may remove this content at anytime! Save it as PDF for later use!

More from @quaesita

Cassie Kozyrkov

@quaesita

15 Jan

1/🧵 Good #DataScience advice that breaks pretty much every rule you learned in class... a thread. (+full blog post linked)

English version: bit.ly/quaesita_rethi…
Spanish version: bit.ly/spanish_rethin…

#AI #MachineLearning #Statistics #RStats #agile

https://t.co/dG4l6vPFBT

2/🧵 Allow your approach to be sloppy at first and burn some of your initial time, energy, and data on informing a good direction later. That's right, you're supposed to start sloppily ON PURPOSE.

3/🧵 Have a phase where the only result you’re after is *an idea of how to design your ultimate approach better.*

Read 16 tweets

Cassie Kozyrkov

@quaesita

14 Aug 20

What makes an analyst excellent? Speed! But that's not as simple as it sounds. They need 6 kinds of speed... 1/🧵

#DataScience #Analytics #rstats #DataAnalytics

bit.ly/quaesita_speed

1) Speed of getting data that’s relevant. (Domain knowledge.)

2) Speed of getting data ready for manipulation. (Software skills.)

Read 15 tweets

Cassie Kozyrkov

@quaesita

22 Apr 20

0/ Essential philosophy for #DataScience, a thread of 32 questions.

Grab a friend (virtually) and tackle these 32 essential questions (all with more than one reasonable answer) that every serious #data professional should answer for themselves.

#philosophy #rstats #statistics

1/ Is it possible to know anything at all?

2/ What does it mean to "know" something?

Read 34 tweets

Cassie Kozyrkov

@quaesita

12 Mar 20

1/10 How to apply the advice of a piece I wrote two years ago to your #COVID19 decision-making:

bit.ly/quaesita_inspi…

2/10 It's not helpful to form an uninformed opinion & then look for media that confirms your views. You'll find it, maybe you'll feel better, but you may as well skip it - it's a waste of time. You already know you'll just confirm whatever you wanted to believe. Instead, do this:

3/10 Think carefully about your ethics/values/goals (for world, community, family, self). (This post isn't about telling you how much of a nice person to be, only how to make decisions within your own moral framework... but I hope you'll choose to be nice.)

Read 11 tweets

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Share this page!

Cassie Kozyrkov

Try unrolling a thread yourself!

More from @quaesita

Cassie Kozyrkov

Cassie Kozyrkov

Cassie Kozyrkov

Cassie Kozyrkov

Did Thread Reader help you today?

Like this author's thread?