Sign in

Chelsea Parlett

@chelseaparlett.bsky.social
12K followers 361 following 1.5K posts

Ph.D, stats lover/writer✍🏼, #statistics #scicomm #datascience #statstiktok 👩🏻‍💻 she/her

PostsRepliesMedia
Chelsea Parlett @chelseaparlett.bsky.social · 20h
Marketing’s incremental impact be like:
A frog with a sword and a princess hat says “never estimate me”
2220
Reposted by Chelsea Parlett
michael 📈 👨‍💻 @mikedecr.computer · 08/10/2026
the couchy distribution
714810
Chelsea Parlett @chelseaparlett.bsky.social · 05/10/2026
My vector when the matrix I multiply it with is not full rank 😕
static.klipy.com
Squidward Flattened by Boulder in Spongebob
ALT: Squidward Flattened by Boulder in Spongebob
1270
Chelsea Parlett @chelseaparlett.bsky.social · 04/10/2026
Me before I knew anything about marketing measurement
Lucille Bluth from Arrested development saying “it’s one conversion Michael, what could it cost, ten dollars?”
2200
Reposted by Chelsea Parlett
Sam Power @spmontecarlo.bsky.social · 03/10/2026
Very good from @timhenke.bsky.social
24215
Chelsea Parlett @chelseaparlett.bsky.social · 03/10/2026
me when all my models assume paid marketing ROI >= 0 but I purposefully avoid brands whose ads really annoy me
static.klipy.com
Shrek Reaction
ALT: Shrek Reaction
4221
Chelsea Parlett @chelseaparlett.bsky.social · 03/10/2026
Me when I'm the latent space of a variational autoencoder 🤪✌️
static.klipy.com
Catriona: I am so random
ALT: Catriona: I am so random
0140
Chelsea Parlett @chelseaparlett.bsky.social · 03/10/2026
What could be BETA than working with some of my faves?? @jgeller1phd.bsky.social is truly the GOAT of explaining beta regression 🐐 I make all my grad students read this when we do GLMs in class🙏
1120
Chelsea Parlett @chelseaparlett.bsky.social · 29/09/2026
As if I needed another reason to hate AI slip posters/ads, AI generated images of food trigger my trypophobia🤢🕳️🕳️🕳️
090
Chelsea Parlett @chelseaparlett.bsky.social · 23/09/2026
I am so intrigued by exactly what about @betanalpha.bsky.social’s office hours next month will be spooky 👻 violated assumptions?? Ghost eigenvalues??? Divergent transitions????
A screenshot of a YouTube upcoming event for Michael Betancourt’s Spooky Office Hours on October 28th
091
Chelsea Parlett @chelseaparlett.bsky.social · 22/09/2026
Sometimes it takes me a moment to remember Suzanne Collins and Susan Collins are different people, which leads to a lot of confusion over headlines 🤔
080
Reposted by Chelsea Parlett
tiddlydump.bsky.social @tiddlydump.bsky.social · 21/09/2026
I love getting to dust this one off
Mac, IASIP: "Well, first of all, everything is a mixture of latent gaussians, so jot that down"
1222
Chelsea Parlett @chelseaparlett.bsky.social · 21/09/2026
Why bell shaped, hmm? Why not ant-hill shaped 🐜? Camel hump shaped 🐪? Turtle silhouette from front on shaped 🐢? Tulip in bloom shaped 🌷?
2253
Chelsea Parlett @chelseaparlett.bsky.social · 18/09/2026
Just got a very exciting book delivery! @grimalkina.bsky.social What an honor it’s been to see Cat’s work over the past 10 years, I’m excited to read this and apply what I learn🤩
The cover of the book “the psychology of Software teams” by Cat Hicks
4162
Chelsea Parlett @chelseaparlett.bsky.social · 16/09/2026
SYMMETRICAL FORMS ARE A BALM TO THE SPIRIT🙏
A drawing of an animal in a beautiful garden with the caption “symmetrical forms are a balm to the spirit”
090
Chelsea Parlett @chelseaparlett.bsky.social · 14/09/2026
sometimes when i'm writing technical material I think to myself, "imagine saying this to a victorian child" and bask in the absurdity that is statistics.
static.klipy.com
Victorian Child Painting
ALT: Victorian Child Painting
1130
Chelsea Parlett @chelseaparlett.bsky.social · 11/09/2026
Do you want a rough answer to your actual question or an elegant answer to a completely different one?
static.klipy.com
Matrix: Red Pill or Blue Pill? What Will You Choose?
ALT: Matrix: Red Pill or Blue Pill? What Will You Choose?
0180
Chelsea Parlett @chelseaparlett.bsky.social · 10/09/2026
This is the way the world ends This is the way the world ends This is the way the world ends Not with a bang but a “you’re right to push back on this”.
2615
Chelsea Parlett @chelseaparlett.bsky.social · 07/09/2026
god watching any YouTube video means you’re inundated with betting platforms and useless (to me) AI tools 🫩 feels like a hopeless glimpse of where our economy is right now.
6120
Chelsea Parlett @chelseaparlett.bsky.social · 05/09/2026
feels a bit passive aggressive, no? 🤨
an email from Audible whos subject reads "Nice! You're 0% complete. Pick up where you left off"
2181
Reposted by Chelsea Parlett
They Might Be Giants @tmbg.bsky.social · 05/09/2026
182137172678
Chelsea Parlett @chelseaparlett.bsky.social · 04/09/2026
"I only see a single number on this entire slide. Adding letters is where you've gone wrong with math" -- @tonypelleriti.bsky.social looking over my shoulder as a I prep a lecture.
a slide showing an Autoregressive model. it discusses the stationarity of a time series.
3190
Chelsea Parlett @chelseaparlett.bsky.social · 01/09/2026
My little ‘Nalysis Of VAriance 🫶🐶
1321
Chelsea Parlett @chelseaparlett.bsky.social · 31/08/2026
I'm writing a book chapter on something and need to brush up because I don't remember *all* the technical details.......BUT YOU KNOW WHO DOES?! ~2023 Chelsea who put all her lectures and notes up on github on the topic
static.klipy.com
Vector Says Oh Yeah!
ALT: Vector Says Oh Yeah!
1260
Chelsea Parlett @chelseaparlett.bsky.social · 31/08/2026
I really like this definition of KL Divergence I use in the math review for my course. It highlights that we can measure the similarity between two distributions by understanding how much information we lose when approximating one distribution with the other.
screenshot of my slide on the KL Divergence in a math review that shows that KL = Cross Entropy - Entropy
1494
Chelsea Parlett @chelseaparlett.bsky.social · 31/08/2026
Do you love providing useful outputs, or do you love building things? In an era of DS absolutely SATURATED with AI use, I’ve been thinking a lot about what I actually liked about being a data scientist. Especially as someone who’s purposefully stayed on the IC track.
2271
Chelsea Parlett @chelseaparlett.bsky.social · 28/08/2026
shot, chaser
The book “the jungle” by uptown SinclairThe book “barons: money, power and the corruption of americas food industry”
181
Chelsea Parlett @chelseaparlett.bsky.social · 25/08/2026
I brought nova to the class I taught tonight. In the middle of one of the peer discussions I hear a quiet “Dr. Parlett…nova…left” 💀💀 Miss nova decided she wanted to explore the science building on her own while we discussed Causal Inference 😂
0221
Chelsea Parlett @chelseaparlett.bsky.social · 24/08/2026
Just saw a tiktok discussing a Marxist analysis of fitness and how fitness is like the one guaranteed place where you’ll enjoy the benefits of your labor in direct proportion to that labor. And…I’m absolutely more motivated to work out now 🏃‍♀️
3262
Chelsea Parlett @chelseaparlett.bsky.social · 18/08/2026
The world does not need more AI generated p-value explainer graphics😅
2180
Chelsea Parlett @chelseaparlett.bsky.social · 15/08/2026
They’re SUCH GOOD BOOKS omg. Check them out ASAP
1244
Chelsea Parlett @chelseaparlett.bsky.social · 11/08/2026
“It kept demanding I also install other increasingly rare and venomous snakes before it would run” 💀💀💀💀💀💀💀🐍
1535
Chelsea Parlett @chelseaparlett.bsky.social · 11/08/2026
I just have a feeling that in my mid 30s I’m gonna get into: - pattern drafting - book binding I’m not there yet, but it’s coming for me.
1150
Reposted by Chelsea Parlett
Danielle Navarro @djnavarro.net · 06/08/2026
The title is a little pointed, but I am very serious here. Academics, medical researchers, and other organisations should not collect data on birth sex. It is rarely relevant, and it almost always leads to privacy violations for transgender people. Please share. I would like people to read this one.
blog.djnavarro.net
Separating sex and gender is statistical malpractice – Notes from a data witch
Statisticians and academics in Australia should ignore the terrible advice of most LGBTIQA+ organisations and read this post instead. The 2020 ABS standard is almost never the correct way to collect d...
14236119
Chelsea Parlett @chelseaparlett.bsky.social · 05/08/2026
I think we should all ask this more. Not just about AI, about a lot of things. What actually WOULD happen if we didn’t do that thing? 🤷‍♀️
1121
Chelsea Parlett @chelseaparlett.bsky.social · 04/08/2026
It’s a rat race out there…and the rats are winnin’
060
Chelsea Parlett @chelseaparlett.bsky.social · 03/08/2026
The good news is: STATISTICS is only hard if you want to do it accurately ! 😅😭😂🫠
2503
Chelsea Parlett @chelseaparlett.bsky.social · 03/08/2026
The good news is: Marketing measurement is only hard if you want to do it accurately ! 😅😭😂🫠
181
Chelsea Parlett @chelseaparlett.bsky.social · 30/07/2026
I have convinced 4 of my friends to start Dungeon Crawler Carl this month 😈 Im re-reading it with them. I’m feeling emboldened so if I start bullying you to read DCC, this is why.
All 8 dungeon crawler Carl books displayed on my bookshelf
5330
Chelsea Parlett @chelseaparlett.bsky.social · 29/07/2026
There are genuinely no words to express how much I LOVE my public library and how HAPPY I am it exists and that some of my tax dollars go towards it🥹 What a beautiful thing to have not only books for FREE but a place to exist without paying for anything, a range of services, a place to meet! 🥰📚
static.klipy.com
Fluffy Persian Cat in a Library
ALT: Fluffy Persian Cat in a Library
2244
Chelsea Parlett @chelseaparlett.bsky.social · 28/07/2026
Whenever I’m having a stressful day I just think about Caleb Hearon pretending to sitting down while already sitting down.
141
Chelsea Parlett @chelseaparlett.bsky.social · 28/07/2026
🦀📈🍑
Three identical Spider-Man pointing at each other. One is labeled “everything is crab” one is labeled “Everything is linear regression, one is labeled “Everything is anus logo”
57114
Chelsea Parlett @chelseaparlett.bsky.social · 28/07/2026
Nothing irrationally irritates me more than an LLM using the word "sharp" to describe everything. My question is not "sharp", I just have human level intelligence and context.
static.klipy.com
Cartoon Raccoon's Disgusted Face
ALT: Cartoon Raccoon's Disgusted Face
1110
Reposted by Chelsea Parlett
Demetri @phdemetri.bsky.social · 28/07/2026
More thoughts about the normal prior vs the student's t prior dpananos.github.io/posts/2026-0...
dpananos.github.io
Adaptive Shrinkage and the Student’s t Prior – Demetri Pananos Ph.D
5216
Chelsea Parlett @chelseaparlett.bsky.social · 26/07/2026
The person in front of me at Costco this week had MULTIPLE bags of Taylor Farms Salad🥬🦠😬
static.klipy.com
Darkwing Duck: Let's Get Dangerous!
Alt: Darkwing Duck: Let's Get Dangerous!
1141
Chelsea Parlett @chelseaparlett.bsky.social · 24/07/2026
I started working with a bunch of Australians and SUDDENLY I can’t stop saying I’m “keen” to do things
static.klipy.com
Kangaroo in Australia
ALT: Kangaroo in Australia
4290
Chelsea Parlett @chelseaparlett.bsky.social · 21/07/2026
A new BOMBshell has entered the villa 🥰 Literally any time I watch Love Island all I can think about is how badly they need a villa corgi (I have a volunteer 👀)
A corgi in a red bucket had and red bandana standing in a lawn
0201
Chelsea Parlett @chelseaparlett.bsky.social · 13/07/2026
I’ve been feeling a much stronger pull towards physical media this year. Between Sony discontinuing physical games, only getting to “license” ebooks that I buy…it just feels BAD. I want to be able to own and LEND my physical media. I want to not worry about it being pulled or censored.
static.klipy.com
Stack of Books on a Table
ALT: Stack of Books on a Table
2171
Chelsea Parlett @chelseaparlett.bsky.social · 09/07/2026
I spoke to Mitch McConnell this morning for 20 minutes and he told me that it’s actually fine to say there’s a 95% chance that your confidence interval contains the population mean 🤗
4443
Chelsea Parlett @chelseaparlett.bsky.social · 07/07/2026
Googling meta-analysis related questions in a marketing context is super annoying because it often assumes you want an analysis of Meta (the company/platform) data 🫠
3200