Justin L
LightNews LightNews
banner
drjaylos.bsky.social
Justin L
@drjaylos.bsky.social
220 followers 500 following 4 posts
Post-doctoral fellow | Musculoskeletal Health Epidemiologist | Sports Physio | Prefer dogs over humans.

#Epidemiology #EpiSky #CausalInference #MSKHealth #SportsMedicine #ProDog
Posts Replies Media Videos
Reposted by Justin L
causalab.bsky.social
Risk difference, risk ratio, win ratio.

Julie Josse, Senior Researcher at INRIA, presents next for the CAUSALab Methods Series @ki.se.

📆 March 18, 2025
⏰ 15.00 CEST/10.00 ET
📍 All Methods Series talks virtual

Register to attend 👇
stats.sender.net/forms/e7JD1d...
#causalinference #publichealth
Julie Josse March 18 Methods Series talk, "Risk Difference, Risk Ratio, Win Ratio: Key Properties for Transportability and Federated Causal Inference."
March 4, 2025 at 10:51 PM
Reposted by Justin L
joshtpm.bsky.social
Hilarious. After the DOGErs set up the new government-wide all employee email system in late January OPM, where it is housed, sent out an advisory which states explicitly that no employee ever has to respond to any of the emails. See 4.2 and 4.3. www.opm.gov/media/kfpozk...
February 23, 2025 at 2:42 PM
Reposted by Justin L
dingdingpeng.the100.ci
Are you interested in age or cohort or period effects? I have just updated my primer on the topic!
Now with a new framing, a handy flow chart to determine which approach may be suitable, and a new empirical example. What do people in Germany think about mothers who work? osf.io/preprints/ps...
Flow chart: Different ways to deal with the age-period-cohort identification problem. Figure with multiple panels showing mean scores over age or survey year, split by cohort, survey year or age. Figure with a canonical solution line and nonlinearities extracted by the bounding analysis approach by Fosse & Winship. Figure showing the resulting bounds on the age, period, and cohort effects.
January 28, 2025 at 2:48 PM
Reposted by Justin L
When estimating a treatment effect with a cluster design, you need to include varying slopes, even if the fit gives warning messages.
statmodeling.stat.columbia.edu/2025/01/23/s...
When estimating a treatment effect with a cluster design, you need to include varying slopes, even if the fit gives warning messages. | Statistical Modeling, Causal Inference, and Social Science
statmodeling.stat.columbia.edu
January 23, 2025 at 2:54 PM
Reposted by Justin L
miguelhernan.org
Upgrade your #causalinference arsenal.

A revision of our book "Causal Inference: What If" is available at miguelhernan.org/whatifbook

Thanks to everyone who suggested improvements, reported typos, and proposed new citations and material.

Enjoy the #WhatIfBook plus code and data. Also, it's free.
December 23, 2024 at 9:28 AM
Reposted by Justin L
markhamillofficial.bsky.social
Posted this at the old place, but thought folks on @bsky.app might like to know I finally met my "mother".
December 12, 2024 at 10:56 PM
Reposted by Justin L
rpsychologist.com
Introducing PowerLMM.js!

A new tool for power analysis of longitudinal linear mixed-effects models (LMMs) – with support for missing data, plus non-inferiority and equivalence tests.

powerlmmjs.rpsychologist.com

Would really appreciate your feedback as I refine this app! Details below 🧵👇
December 11, 2024 at 10:20 AM
Reposted by Justin L
richarddriley.bsky.social
Christmas 2024 gifts for your 'significant' other 🎄

1. How to get your article rejected by the BMJ (at Christmas or anytime)

Video: www.youtube.com/watch?v=iu4V...

Article: www.bmj.com/content/379/...
How to get your article rejected by the BMJ: 12 common statistical issues
YouTube video by Richard_D_Riley
www.youtube.com
December 11, 2024 at 8:34 AM
Reposted by Justin L
gscollins.bsky.social
It's that time of year to bring out our @bmj.com xmas paper On the 12th Day of Christmas, a Statistician Sent to Me... from 2022 (w/ @richarddriley.bsky.social) with our list of commonly seen statistical faux pas🎄

www.bmj.com/content/379/...

#StatsSky #EpiSky
December 5, 2024 at 3:00 PM
Reposted by Justin L
pwgtennant.bsky.social
STOP IGNORING SELECTION BIAS!
#EpiSky #StatsSky
Tiger Woods and John Daly meme. Where Tiger Woods - looking professional - is labelled "doubly-robust control for confounding". And John Daly - looking very relaxed and smoking a cigarette - is labelled "Selection bias".
December 4, 2024 at 12:08 PM
Reposted by Justin L
andrew.heiss.phd
Woohoo, our paper is live!

This haiku summarizes it best:

Raw model results?
Stop! Hard to understand! Use
{marginaleffects}
vincentab.bsky.social Vincent Arel-Bundock @vincentab.bsky.social · Dec 2
Our JSS article is out!

And now I get to focus on {marginaleffects} 1.0.0. Stay tuned.

www.jstatsoft.org/article/view...
December 2, 2024 at 9:09 PM
Reposted by Justin L
umhoops.com
You need Peacock to watch Michigan basketball this season -- starting next week!

Peacock just raised prices BUT is having its best sale of the year. You can get a year's subscription for just $19.99 this week (75% off).

Sign up here: imp.i305175.net/c/2991622/22...
Peacock: Stream TV and Movies Online, Watch Live News and Sports
Watch TV shows and movies online with Peacock. Stream iconic shows and movies, exclusive Peacock Originals, live news and sports and more.
imp.i305175.net
November 25, 2024 at 5:20 PM
Reposted by Justin L
conjugateprior.org
The organisers have sensibly kept this hidden as long as possible, but here's the talk.

youtu.be/m56YEkkWYzI

CW dad jokes, tendentious claims, Monty Hall problem.

Best watched at zero speed, while doing something actually productive instead.
BEMC NOV 2024 - William Lowe - Collider bias (complimentary)
YouTube video by Berlin Epidemiological Methods Colloquium
youtu.be
November 16, 2024 at 10:14 AM
Reposted by Justin L
causalab.bsky.social
Save the dates!

CAUSALab’s Summer Courses on #causalinference return June 2025. Detailed course information to be announced in the coming weeks.

Join our 2024-2025 course listserv to receive updates first: harvard.az1.qualtrics.com/jfe/form/SV_...

Learn more:
causalab.hsph.harvard.edu/courses/
Save the Dates! CAUSALab Summer Courses on Causal Inference Week 1 Courses: June 16-20, 2025 No courses on June 19th in observance of Juneteenth. June 20th extended to full day. Week 2 Courses: June 23-27, 2025
November 26, 2024 at 6:05 PM
Reposted by Justin L
statsepi.bsky.social
As well as prognostic factors that *aren't* imbalanced! It's amazing. Links to good advice on "how" in the post below.
November 26, 2024 at 2:47 PM
Reposted by Justin L
andrew.heiss.phd
From the archives: Have you (like me!) wondered what the ATT means and how it's different from average treatment effects? I use #rstats to explore why we care about (and how to calculate) the ATE, ATT, and ATU #polisky #episky #econsky www.andrewheiss.com/blog/2024/03...
Mirrored histogram showing “weird” parts of the population: treated people who were unlikely to be treated, and untreated people who were likely to be treated Mirrored histogram showing pseudo-populations of treated and untreated people that have been reweighted to be more comparable and unconfounded Table showing potential and realized outcomes for 9 simulated people Before we calculate these different treatment effects with the realized outcomes instead of the hypothetical potential outcomes, let's look really quick at the practical difference between the true ATE, AT 1, and ATU. All three estimands are useful for policymaking! The ATE is -15, implying that mosquito nets cause a 15 point reduction in malaria risk for every person in the country. This includes people who live at high elevations where mosquitoes don't live, people who live near mosquito-infested swamps, people who are rich enough to buy Bill Gates's mosquito laser, and people who can't afford a net but would really like to use one. If we worked in the Ministry of Health and wanted to know if we should make a new national program that gave everyone a free bed net, the overall reduction in risk is -15, which is probably pretty good! The ATT is -16.29, which is bigger than the ATE. The effect of net usage is bigger for people who are already using the nets. This is because of underlying systematic reasons, or selection bias. Those using nets want to use them because they need them more or can access them more easily-they might live in areas more prone to mosquitoes, or they can afford to buy their own nets, or something else. They know themselves and understand some notion of their personal individual causal effect and seek out nets. If we removed access to their nets, it would have a strong effect. The ATU is -13.63, which is smaller than the ATE. The effect of net usage is smaller for people who aren't using the nets. Again, this is because of selection bias. Those not using nets are likely not using them for systematic reasons-they live far away from mosquitoes, they've received a future malaria vaccine, they have some other form of mosquito abatement, or something else. Because they can read their own minds, they know that mosquito net use won't do much for them personally, so they don't seek out nets. If we expanded access to nets to them, they wouldn't benefit
November 22, 2024 at 2:51 PM
Reposted by Justin L
f2harrell.bsky.social
#statistics thought of the day:For measurement agreement/observer variability, consider powerful & easier to interpret (than variance components) all-relevant-pairs-mean-absolute-discrepancies, "relevant" chosen to match inter/intra observer interest hbiostat.org/bbr/obsvar #StatsSky #epiSky
Biostatistics for Biomedical Research - 16  Analysis of Observer Variability and Measurement Agreement
hbiostat.org
November 19, 2024 at 4:25 PM
Reposted by Justin L
yegphysio.bsky.social
no idea what i’m doing on here yet but these starter packs seem cool. let’s put one together for the SEM crowd, yes?

please add folks and grow the community!

go.bsky.app/VYkpXdS
November 15, 2024 at 8:58 PM
drjaylos.bsky.social
Getting ready for winter adventures. #dogsofbluesky #adventuredogs
November 19, 2024 at 4:56 AM
Reposted by Justin L
rmcelreath.bsky.social
Was recommending to a colleague POSTSTRATIFICATION (*distant thunder*) and even MULTILEVEL REGRESSION AND POSTSTRATIFICATION (*closer thunder*). I pointed them at @rohanalexander.bsky.social's book, chapter 16, which is free online. Contains an election forecasting example.
Telling Stories with Data - 16  Multilevel regression with post-stratification
tellingstorieswithdata.com
November 14, 2024 at 12:42 PM
Reposted by Justin L
andrew.heiss.phd
Had a blast teaching this today! All the #rstats materials are available at talks.andrewheiss.com/2024-11-13_u...
Go beyond OLS! An introduction to Poisson, Beta, and zero-inflated Beta Bayesian distributional regression While ordinary least squares (OLS) regression is an important method for modeling relationships, it’s not always well-suited for outcome variables that are proportions (bounded between 0–100%) and counts (limited to whole numbers). In this workshop, you’ll learn how to use Bayesian regression techniques to get richer and more detailed information from more specialized types of data. Using R, Stan, {brms}, and {marginaleffects} you’ll learn how to (1) correctly model counts and proportions and (2) process, plot, interpret, and communicate the results. Author Andrew Heiss Affiliation Georgia State University Published November 13, 2024 Workshop presented at the Département de science politique at the Université de Montréal Plot showing the predicted count of anti-LBGT-discrimination laws across a range of urbanness in Democratic, Republican, and swing states, using Bayesian Poisson regression Plot showing the predicted proportion of protected-at-birth tetanus vaccination rates across a range of GDP per capita and 6 different global regions, with vertical lines showing the slopes of the predicted lines for some typical/hypothetical countries, all using Bayesian Beta regression Plot showing multiple Beta distributions across two different parameterizations (shapes and µ+φ)
November 13, 2024 at 10:08 PM
Reposted by Justin L
statsepi.bsky.social
In clinical research, you will often receive feedback on study design, stats, and/or data analysis from an editor or reviewer that is simply wrong. Here is a list of common "statistical myths" and references you can use to push back.

discourse.datamethods.org/t/reference-...
Reference Collection to push back against "Common Statistical Myths"
Note: This topic is a wiki, meaning that this main body of the topic can be edited by others. Use the Reply button only to post questions or comments about material contained in the body, or to sugge...
discourse.datamethods.org
November 12, 2024 at 6:19 AM
Reposted by Justin L
f2harrell.bsky.social
#statistics thought of the day:Is an EHR randomization defective? Simple example contrasting frequentist and Bayesian inference. Would you rather have Pr(getting MORE heads than what we observed if a coin is fair),or Pr(Pr(heads) is outside [0.48, 0.52])? #stats #rstats hbiostat.org/bayes/bet/ev...
3  Measures of Evidence – Introduction to Bayes for Evaluating Treatments
hbiostat.org
November 12, 2024 at 12:30 PM
Reposted by Justin L
retractionwatch.com
Researchers find nearly 30% of papers “that stated significances (or their absence) are based on the presence of a single influential data point.”
Scientific reasoning driven by influential data: resuscitate dfstat!
In biomedical literature, one of the most widely employed statistical procedures to analyze and visualize the association between two variables is linear regression. Data points that exert influence o...
www.biorxiv.org
November 8, 2024 at 10:29 PM