Rohan Alexander
@RohanAlexander
Followers
3,485
Following
1,528
Media
568
Statuses
2,746
I spend too much money on books. And too much time writing them. 🇦🇺.
Joined March 2009
Don't wanna be here?
Send us removal request.
Explore trending content on Musk Viewer
#今月描いた絵を晒そう
• 180721 Tweets
Mombasa
• 73780 Tweets
Griezmann
• 57232 Tweets
#JusticeForBruce
• 55307 Tweets
#CDTVライブライブ
• 44882 Tweets
PASTEL FRESHTORY X GF
• 25737 Tweets
OUROAD GRAND OPEN HOUSE
• 18731 Tweets
Grizou
• 17820 Tweets
おはスタ
• 17412 Tweets
シャウエッセン
• 16084 Tweets
#chibalotte
• 15190 Tweets
Deschamps
• 14657 Tweets
フルサイズ
• 13662 Tweets
OHMLENG IDEA LIVE
• 11390 Tweets
Pinned Tweet
My book about introductory data science was published today. It is available online - - and you can buy a print copy: . I would appreciate your feedback and suggestions for improvements.
17
147
630
Peer review is broken. Reason
#912
:
Two reviewers both declined, and each suggested each other.
31
31
822
US reviewers: EXPLAIN TO ME WHY WE SHOULD CARE ABOUT AUSTRALIA/CANADA/UK?
Also US reviewers: I wrote a paper about Rhode Island state legislative districts that you didn't cite.
7
73
692
Can we all agree to try to normalize 25 min (not 60 min) meetings in academia?
30
31
656
When grad students want to learn something they take a course on it. When professors want to learn something they teach a course on it.
13
39
546
OMFG it exists (photo from a colleague in the UK). I apologize in advance, but this account is going to be (even more) insufferable for the next month.
21
34
449
Economists, please, just read, like, one book on how to code.
16
28
435
We all just become the weighted average of our supervisors.
15
23
394
A hill I will die on: Every stats student needs a SQL course; every data science student needs a history of stats course.
16
21
332
Gelman (!) wrote a nice blurb for my book; I think I'd be okay if that's all I accomplish as an academic.
9
5
247
Happily this week I get to cover my favourite sentence in all my teaching notes.
8
27
232
This happens more commonly than you may think. My stats undergrad classes have found two without even trying in recent econ pubs. Hard to believe a paper unless all the code is there these days.
Thinking about kinds of analysis errors. There are errors of design, doing the wrong analysis intentiona, & also errors of execution, doing the wrong analysis by accident. Like this 2016 example in which country codes entered as continuous by accident.
6
24
180
8
28
196
Comparison of my email style:
In 🇺🇸:
Dear Professor SURNAME,
My paper is the greatest & you’re the best prof... Want to email?
In 🇨🇦:
Dear Professor SURNAME,
My paper does X... Want to get coffee?
In 🇦🇺:
Hey Bob,
My paper is wrong in 27 ways... Want to get a beer?
4
31
191
My boss doesn’t have write access. I’m NEVER changing this.
7
10
178
Re-drafting Telling Stories with Data:
Feedback/typos/opinions/etc greatly appreciated.
4
25
171
My advice is always the same - if you're in a quant social sciences degree, the best course you could take is to make a reading group and go through these.
The 2023 edition of my long-running anti-establishment art-science-fusion code-therapy smooth-baritone causal inference & Bayesian data analysis course is complete. 20 lectures, from the basics of causal inference & Bayesian updating to mixed models & Gaussian processes. 1/2
48
477
2K
2
21
168
Can people please stop innovating for just like a year so I can finish my book and have it be relevant for at least a few months.
9
7
159
Scientific progress is cool. Five years ago, topic models of some hansard took a year of my life, and today it's literally just a ten-line textbook example that students will probably see as trivial.
2
7
147
The Toronto Data Workshop on Reproducibility will be held online on 25/26 February 2021 (two full days!). Free. All welcome.
Register:
Details:
Keynotes:
@evavivalt
(practices)
@minebocek
(teaching)
@rianaminocher
(evaluating).
1
77
142
Stats textbooks: "here's a few toys, go for it".
Stats research papers: "DO NOT TAKE THAT TOY OUT OF THE PLAYGROUND. DO NOT MIX THAT TOY WITH OTHER TOYS. THAT TOY CAN ONLY BE USED ON A TUESDAY".
1
18
134
Thanks to
@crcgrubbsd
for the free book. Highly recommend being a reviewer for CRC - it's like seeing a year or two into the future & you get free books!
6
11
124
If you want to know why I'm so dubious of everyone else's code, it's because I know the state of my own
4
9
116
Them: What are your contributions to your department?
Me:
3
4
116
Student survey feedback: I like how you just wing it in lectures.
Dear reader, I literally spent 12 hours yesterday preparing 30 minutes of material.🧑💻
3
0
113
Does anyone have opinions about what a ‘history of stats’ course should look like? If so, would love to hear them or see some examples. Probably grads, or upper undergrads. My current thinking here:
31
19
105
Starting tonight
@monjalexander
and I are going to watch the McElreath lectures. Average 1.5 each day, for two weeks. Anyone want to match us to keep us honest?
5
5
105
Got this approved in time for winter 2023. It's a reimagined first year quant methods course integrating reproducibility, ethics, workflow, writing, & portfolio building. Any suggestions?
5
13
101
Today in office hours:
Undergrad reproducing a paper: ‘What do I do when there's a bunch of small unexplained decisions? The sign changes if you use the full sample instead of the weird subset they use in the paper.'
Welcome to the reproducibility crisis my sweet summer child.
1
7
98
The Toronto Data Workshop on Reproducibility will be held over two full days via Zoom on 25/26 February 2021.
Free. All welcome.
Registration:
Details:
Keynotes by:
@evavivalt
@minebocek
@rianaminocher
3
60
99
Despite the dumpster fire of the past 18 months, 2 children, 3 countries (and approx 10 cities),
@monjalexander
has published a paper every year since 2013!
6
1
95
If you’re not giving your data science students the chance to make a website and fill it with projects, then my
@UofTStatSci
students are taking their jobs ;)
5
4
97
U of T
@UofT_dlsph
is advertising three tenure stream assistant professor faculty positions in Data Science Methods for Population Health and Health Systems:
We're hoping for a wide reach. Please RT and apply.
1
89
95
Monica's possibly the most important and thoughtful applied statistician of her generation right now.
I wrote a post about applied statistics, data, and answering hard questions. The takeaway is that when doing and teaching applied statistics, we need to emphasize engaging with data, and understanding data in context.
3
50
261
4
3
89
Some of the papers written by
@UofTStatSci
undergrads during winter class I taught. They received little guidance from me apart from "do original research in a completely reproducible way". The kids are alright.
#RStats
2
17
91
Plan, as at Nov 2017:
- Get data into usable form (1 week)
- Clean data (1 month)
- Analyse data (2 months)
- Write up paper (1 month)
Reality, as at July 2018:
- Get data into usable form (8 months)
#phdlife
2
8
88
"Probably the best book of the year."
Going to tell my Mum about that one!
👉 Probably the best book of the year.
👉 A must-read.
👉 Check it out for FREE on
1
3
11
1
8
87
Unclear to me that our marriage will survive Monica’s markups to my book.
5
0
89
New office. Three empty bookcases. Challenge accepted.
4
0
85
I gave a presentation, and someone asked something that required unprepared live coding, and the code worked!
Good day to buy a lottery ticket.
3
1
83
3yo: Daddy, what your favourite shape?
Me: Hexagon
3yo: But why?
Me: While we could start in 1976 in Bell Labs or even the 1970s at the MIT Media Labs, we will begin our story in 1991 in New Zealand when the statisticians Ross Ihaka and Robert Gentleman….
3yo: *Wanders away*
3
4
85
Monica and I hosted adjacent workshops, and the baby slept through the handover. Going to buy a lottery ticket.
2
0
82
@PBlanchenay
I think it would be better if they used a project, rather than rely on users changing a file path, because that way the code could be more easily run by computers. And spaces in the file names should be avoided. I found this paper really interesting:
2
5
82
Reminder that the Journal of Statistics and Data Science Education has a special issue on teaching reproducibility and workflows.
Please submit!
Due 15 Sept.
Details:
2
30
77
Does anyone else take like (at least) three iterations of teaching a course before the pieces start to come together and the edges smooth over?
8
1
78
Is the best value thing you can do with one minute to send a thank you note to someone? I just got one, literally just a four line email, and yet my day is totally made.
2
3
78
I’m sorry Outlook, but in this house we embrace uncertainty.
3
3
78
We define interdisciplinary research to mean that which gets both rejected from NSERC for being too SSHRC, and rejected from SSHRC for being too NSERC.
4
4
77
Explicitly teaching data cleaning today. Anyone done this and have hands-on exercises that they'd recommend?
15
5
72
Excited to get a copy of ‘Data Science: A First Introduction’. Highly recommended :) Congratulations to
@TiffanyTimbers
, Campbell, and Lee (and
@crcgrubbsd
and the CRC team).
1
14
68
On Friday 23 April 1-4pm (Toronto time) I’ll be giving a free introduction to multilevel regression with post-stratification (MRP) in
#rstats
via Zoom. Notes/recording available after. All welcome. Please register here:
0
26
68
Unconfirmed photo of the first draft of my thesis with my supervisors' comments.
4
3
68
Me: My students don’t understand me at all.
My students: Hex sticker card of a duckie-shaped word cloud of Statistical Rethinking, macaroons, and
#RStats
onesies.
#bestnewborngiftsever
1
4
66
Submitted. Thanks you to David, Curtis, Robin, Erica, and Taylor & Francis team. And to Monica for looking after the children (and a last minute notation fix!).
7
1
62
She would never mention this, but
@monjalexander
got an email from a student who said:
‘There aren’t a lot of people in academia who make it seem like academia could be a place for me, but you are one of them!’
Makes it all worth it, doesn’t it.
1
0
63
In case anyone in Toronto is free Thursday afternoon please come to a book launch for Telling Stories with Data.
Edward Alexander (age 4) says "I like the fire on the cover"
Hardcopy is 20% off:
And free online here:
6
17
62
Today the first of an eight-course series on foundational data science skills started at DSI.
This all came together in less than a year - developing curriculum, lesson planning, making slides and assessment, dealing with other deps, getting funds, hiring instructors/TAs, etc.
3
8
61
Toronto Data Workshop brings together academia and industry to share data science best practice. Anyone is welcome to attend—it is free and you don’t need to be affiliated with the university. For an invitation please sign up: .
0
25
59
I've peaked! Saloni included Telling Stories with Data in the brilliant (and free) newsletter "Scientific Discovery"!
3
10
60
Best books I read in 2022:
- Efron & Hastie, "Computer Age Statistical Inference" (🙏 Monica)
- Light, Singer & Willett, "By Design"
- Mindell, "Digital Apollo" (🙏 Dan)
- Morange, "A History of Biology"
3
3
59
With one day to go until teaching starts, the grandparentals have arrived in a (largely) child-care capacity for two months.
3
0
59
Best I can tell Canadian public policy assumes everyone’s been here for generations.
- House? Inherit it!
- Doctor? Use the same one your family used!
- Childcare? Grandparents can help!
- Holiday? Go to your cottage!
4
3
58
I’ve taught this student everything I know. Never been more proud.
2
2
55
OMG a draft of my book exists. Would anyone be open to reading a chapter or two and providing critical comments please?
18
2
56
When life gives you 🍋 [a fourth wave and you bought everything you could possibly need for remote teaching many waves ago, but still got $1K that had to be spent on teaching-related expenditure before the end of the year] make lemonade.
4
0
53
Tres last minute, but I'm giving a 2 hour intro to R session today at 4pm. Four case studies: elections, homelessness, Spotify, and (baby’s nap length pending) genetics.
Free & open to all. For an invite please fill out this form:
Will post recording.
3
12
56
A brief review of forthcoming intro data science textbook - ‘Data Science: A First Introduction’ by Tiffany-Anne Timbers, Trevor Campbell, and Melissa Lee:
3
12
54
Story time is a bit different when your Mum’s a statistics professor.
4
1
54
I surveyed the students in my
@UofTInfoFaculty
class, and as a fun/attention check question asked ‘After COVID ends, if I visited the place you grew up, what should I do that’s not in the guidebooks?’ Some unexpectedly emotion-invoking responses.
0
1
52
The Toronto Workshop on Reproducibility will be held online this Thursday and Friday. Free. All welcome.
Last chance to register:
Details:
Keynotes:
@evavivalt
(practices)
@minebocek
(teaching)
@rianaminocher
(evaluating).
3
28
54
Annie Collins and I look at the reproducibility of COVID-19-related pre-prints and find that it's not great. We download a lot of pre-prints, look for markers of open code/data, and are unable to find them for roughly 3/4.
Details:
Comments welcome!
4
14
53
Proud of Monica for getting rcbayes on CRAN. The demands on her have barely reduced even though she's on mat leave caring for an infant. Amazed at her ability to support students, apply for grants, research, publish, write a textbook, and contribute to her intellectual community.
Our
#rstats
package 'rcbayes' is now on CRAN. The package has functions to calculate, visualize and estimate Rogers-Castro migration schedules using
@mcmc_stan
.
6
39
175
2
3
51
I get to introduce Andrew Gelman at today’s
@UofTDSI
launch 😊🎉!!! My remarks are here:
(Still time to fix them if anyone finds mistakes 😥)
5
2
51
Toronto Data Workshop returns in Winter 2024. All welcome. Please share.
2
24
46
Monica’s off to
@MPIDRnews
as part of a sabbatical award. The children did not fit into carry on, so I’m solo-Dad-ing for the week. Please send 🍷 & ☕️.
3
0
49
Just getting organized for Toronto Data Workshop invites - anyone know of someone doing super interesting work in data science (ideally v. respectful of the data bit), who could use more of a platform?
4
27
47
‘Hey Rohan, can you just do this thing real quick in Python?’
Python: nope.
0
4
49
Me 3 years ago: When I have children, there will not be any special meals.
Me today: The child is eating mini cucumbers with ketchup for dinner, and honestly that’s a win.
1
1
50
Got to do my fav class today - we literally do Fisher's tea tasting experiment. The dean didn't agree to reimburse Wedgwood, maybe next year...
4
3
50
Applications to present at the 2022 Toronto Workshop on Reproducibility are now open (deadline 10 December):
0
19
50
I gave a brief presentation about why it’s important to plot your raw data:
Based on this blog post:
4
4
47
Toronto Data Workshop is back for Fall with a special series of talks featuring
@UofT
speakers on the relationship between data science and their other field of expertise.
This week
@profchapple
from
@UofTCities
. All welcome. Sign up:
0
18
48
Thank you to
@MarijaTaflaga
and
@politicsANU
for funding (and to Monica for solo-parenting) to let me spend two weeks in Canberra fixing my book draft. If anyone's around the ANU and wants to grab coffee let me know.
6
2
45
Incredible way to start the day 🙏😢😱. Still shocked to hear that people other than
@monjalexander
and my Mum read these notes.
Looking for post-stratification (MRP) examples, I stumbled across
@RohanAlexander
's course notes.
THESE ARE FANTASTIC – code, lit review, quizes, memes
Nice synthetic examples with all the code, like this simulated toddler bedtimes example:
1
33
200
1
2
47
Today
@prof_lyons
and I are announcing a call for applications for a $1,000 research stipend that is open to U of T students and recent graduates who identify as a member of a visible minority group, racialized group, or as a person of colour. Details: .
1
31
47
Brilliant new release of marginaleffects - one of the best R packages around.
𝚖𝚊𝚛𝚐𝚒𝚗𝚊𝚕𝚎𝚏𝚏𝚎𝚌𝚝𝚜 0.9.0 is a BIG📦release! Simple yet powerful tools to help you interpret the results of over 70 classes of models in
#RStats
(LM, GLM, GAM, discrete choice, mixed-effects, bayesian, etc.) 🧵 on some cool new stuff.
26
290
2K
2
3
46
Happy Canada Day! 🇨🇦 4 years for Monica and I in Canada 🍁🍁🍁🍁
1
2
46
I flew to give a talk in a different city and the host didn't show up or respond to emails, and no one knew I was coming. They replied to my email 12 hours later saying they were on a fly-out on the other side of the world. Didn't reimburse me.
My worst seminar visit was when I showed up and the first thing I'm told is "Um, so, Erik Lander is visiting today also, and nobody wants to meet with you." I ended up attending Erik Lander's talk, which was fine, though I normally wouldn't fly across the continent just for that.
9
6
159
8
3
45
