Jess Rohmann
@jlrohmann.bsky.social
PhD | Epidemiology, Applied #CausalInference, #PublicHealth, Stroke research, improving quality, peer review, higher ed & research assessment reform
@ Charité in #Berlin
Likes: improving science & improv comedy
#EpiSky #Epidemiology #HigherEd #AcademicSky
@ Charité in #Berlin
Likes: improving science & improv comedy
#EpiSky #Epidemiology #HigherEd #AcademicSky
BERLIN + CAUSAL GRAPHS = HAPPINESS.
Agree?
Follow: bsky.app/profile/appl...
Spoiler: Edition #3 in the making for Q2 2026! Stay tuned!
Agree?
Follow: bsky.app/profile/appl...
Spoiler: Edition #3 in the making for Q2 2026! Stay tuned!
November 12, 2025 at 4:44 PM
BERLIN + CAUSAL GRAPHS = HAPPINESS.
Agree?
Follow: bsky.app/profile/appl...
Spoiler: Edition #3 in the making for Q2 2026! Stay tuned!
Agree?
Follow: bsky.app/profile/appl...
Spoiler: Edition #3 in the making for Q2 2026! Stay tuned!
Here's another photo for the memory box / departmental scrapbook 😀
October 29, 2025 at 10:45 AM
Here's another photo for the memory box / departmental scrapbook 😀
Hey Berlin-based researchers who also wear journal editor hats! 👩🔬
We’re building a cross-discipline network for people like you, and we’d love your help reaching more folks. If you know contacts or places—like groups, mailing lists, or platforms—where we can share our invite, please DM me! Thanks!!
We’re building a cross-discipline network for people like you, and we’d love your help reaching more folks. If you know contacts or places—like groups, mailing lists, or platforms—where we can share our invite, please DM me! Thanks!!
September 29, 2025 at 7:05 AM
Hey Berlin-based researchers who also wear journal editor hats! 👩🔬
We’re building a cross-discipline network for people like you, and we’d love your help reaching more folks. If you know contacts or places—like groups, mailing lists, or platforms—where we can share our invite, please DM me! Thanks!!
We’re building a cross-discipline network for people like you, and we’d love your help reaching more folks. If you know contacts or places—like groups, mailing lists, or platforms—where we can share our invite, please DM me! Thanks!!
No one teaches collider bias like @pwgtennant.bsky.social in the #CausalIntroCourse.
Lighthearted dice rolling 🎲 before the big reveal. Puzzled looks. Then 🤯 shock, as the implications for their research fields sank in. You could hear a pin drop during the compelling lecture that followed! 👏🏻
Lighthearted dice rolling 🎲 before the big reveal. Puzzled looks. Then 🤯 shock, as the implications for their research fields sank in. You could hear a pin drop during the compelling lecture that followed! 👏🏻
July 9, 2025 at 9:32 PM
No one teaches collider bias like @pwgtennant.bsky.social in the #CausalIntroCourse.
Lighthearted dice rolling 🎲 before the big reveal. Puzzled looks. Then 🤯 shock, as the implications for their research fields sank in. You could hear a pin drop during the compelling lecture that followed! 👏🏻
Lighthearted dice rolling 🎲 before the big reveal. Puzzled looks. Then 🤯 shock, as the implications for their research fields sank in. You could hear a pin drop during the compelling lecture that followed! 👏🏻
I love @epidbydesign.bsky.social 's peer review policy for questions @ this morning's #SER2025 session! 🤩 All conferences should do this! @societyforepi.bsky.social
June 12, 2025 at 2:18 PM
I love @epidbydesign.bsky.social 's peer review policy for questions @ this morning's #SER2025 session! 🤩 All conferences should do this! @societyforepi.bsky.social
Enjoying the side by side timelines of T Swift's life vs Epidemiology journal's evolution kicking off the "Eras" session at @societyforepi.bsky.social #SER2025. 😂
June 11, 2025 at 2:41 PM
Enjoying the side by side timelines of T Swift's life vs Epidemiology journal's evolution kicking off the "Eras" session at @societyforepi.bsky.social #SER2025. 😂
Another nice #EuroCIM2025 poster by Christiane Didden. She presents the Oaxaca-Blinder decomposition & how we think about the “explained” part of disparities—specifically in the context of the gender pay gap....
April 14, 2025 at 4:52 PM
Another nice #EuroCIM2025 poster by Christiane Didden. She presents the Oaxaca-Blinder decomposition & how we think about the “explained” part of disparities—specifically in the context of the gender pay gap....
Here's a closer look! Big shout out to @annickdp.bsky.social al, @louisepoppe.bsky.social, @svansteelandt.bsky.social , @pwgtennant.bsky.social , @georgiatomova.bsky.social & many others for putting their heads together on this...
April 14, 2025 at 4:37 PM
Here's a closer look! Big shout out to @annickdp.bsky.social al, @louisepoppe.bsky.social, @svansteelandt.bsky.social , @pwgtennant.bsky.social , @georgiatomova.bsky.social & many others for putting their heads together on this...
Tutorials are most useful when they also address substantive questions of interest. I think #episky also has some good sneak tutorial examples (which often serve as what Julia calls "template articles")! You can find her whole slide deck at osf.io/ujpsq (extra points for openness 🏆)
#EuroCIM2025
#EuroCIM2025
April 10, 2025 at 3:13 PM
Tutorials are most useful when they also address substantive questions of interest. I think #episky also has some good sneak tutorial examples (which often serve as what Julia calls "template articles")! You can find her whole slide deck at osf.io/ujpsq (extra points for openness 🏆)
#EuroCIM2025
#EuroCIM2025
I really agree... Especially because domain subject-matter knowledge is so critical for causal inference. There is often little awareness of what's out there and making "first contact" with a field is so important. #EuroCIM2025
April 10, 2025 at 3:08 PM
I really agree... Especially because domain subject-matter knowledge is so critical for causal inference. There is often little awareness of what's out there and making "first contact" with a field is so important. #EuroCIM2025
You know Pearl's causal ladder, but Julia introduced a different type of ladder : how can we get cutting edge causal inference methods into applications? #EuroCIM2025 so important!
April 10, 2025 at 3:06 PM
You know Pearl's causal ladder, but Julia introduced a different type of ladder : how can we get cutting edge causal inference methods into applications? #EuroCIM2025 so important!
And you can really replace "psych" with most any applied medical research field... We can do better! This slide got a lot of laughs at #EuroCIM2024 (just so true!)
Tough to do at 5pm, but @dingdingpeng.the100.ci is a pro!
Tough to do at 5pm, but @dingdingpeng.the100.ci is a pro!
April 10, 2025 at 2:59 PM
And you can really replace "psych" with most any applied medical research field... We can do better! This slide got a lot of laughs at #EuroCIM2024 (just so true!)
Tough to do at 5pm, but @dingdingpeng.the100.ci is a pro!
Tough to do at 5pm, but @dingdingpeng.the100.ci is a pro!
Final talk of the day... By the fabulous Julia Rohrer aka @dingdingpeng.the100.ci #EuroCIM2025 definitely wins the award for best title slide thus far. (Anyone want to try to top it tomorrow morning??)
April 10, 2025 at 2:54 PM
Final talk of the day... By the fabulous Julia Rohrer aka @dingdingpeng.the100.ci #EuroCIM2025 definitely wins the award for best title slide thus far. (Anyone want to try to top it tomorrow morning??)
Another #EuroCIM2025 poster for #episky:
Elise Dumas presented "Inference on sustained treatment strategies, with a case study on young women with breast cancer" on behalf of colleagues from EPFL and France. Lots of interesting stuff to unpack here!...
Elise Dumas presented "Inference on sustained treatment strategies, with a case study on young women with breast cancer" on behalf of colleagues from EPFL and France. Lots of interesting stuff to unpack here!...
April 10, 2025 at 1:52 PM
Another #EuroCIM2025 poster for #episky:
Elise Dumas presented "Inference on sustained treatment strategies, with a case study on young women with breast cancer" on behalf of colleagues from EPFL and France. Lots of interesting stuff to unpack here!...
Elise Dumas presented "Inference on sustained treatment strategies, with a case study on young women with breast cancer" on behalf of colleagues from EPFL and France. Lots of interesting stuff to unpack here!...
There are some very interesting posters #EuroCIM2025! Sharing a handful here I thought might be of interest to #episky: #1 - "Confounding of the competing event in time-to-event analyses" by Jost Viebrock, Bianca Kollhorst & Vanessa Didelez.
This poster looks at "competing confounding", that is...
This poster looks at "competing confounding", that is...
April 10, 2025 at 12:04 PM
There are some very interesting posters #EuroCIM2025! Sharing a handful here I thought might be of interest to #episky: #1 - "Confounding of the competing event in time-to-event analyses" by Jost Viebrock, Bianca Kollhorst & Vanessa Didelez.
This poster looks at "competing confounding", that is...
This poster looks at "competing confounding", that is...
➕ And an important intro slide for those not yet familiar with bipartite interference!
April 9, 2025 at 9:55 AM
➕ And an important intro slide for those not yet familiar with bipartite interference!
Last three slides (couldn't type fast enough, but still snapped them if you're interested!). I guess I'm rusty on live posting ;) paper will be on arxiv soon!
April 9, 2025 at 9:54 AM
Last three slides (couldn't type fast enough, but still snapped them if you're interested!). I guess I'm rusty on live posting ;) paper will be on arxiv soon!
& the estimands corresponding to the relevant questions. ie. What if we intervene and compare all plants having filters vs. none? What if we compare status quo filter situation vs. none? What about all filters vs. status quo? These are all unique estimands!
April 9, 2025 at 9:54 AM
& the estimands corresponding to the relevant questions. ie. What if we intervene and compare all plants having filters vs. none? What if we compare status quo filter situation vs. none? What about all filters vs. status quo? These are all unique estimands!
Assumptions first! (Of course, this is a causal inference conference)
April 9, 2025 at 9:54 AM
Assumptions first! (Of course, this is a causal inference conference)
Here's more detail about what their forthcoming work adds. After relevant estimands are defined (for policy decision making), they developed design-based estimation overcoming the issues commonly encountered in the presence of bipartite inference.
April 9, 2025 at 9:54 AM
Here's more detail about what their forthcoming work adds. After relevant estimands are defined (for policy decision making), they developed design-based estimation overcoming the issues commonly encountered in the presence of bipartite inference.
I liked her discussion about how there are two separate "tasks" at hand here: defining the structure of the interference and defining the estimands (which are typically not informed by the direct effects typically estimated in these types of studies).
April 9, 2025 at 9:54 AM
I liked her discussion about how there are two separate "tasks" at hand here: defining the structure of the interference and defining the estimands (which are typically not informed by the direct effects typically estimated in these types of studies).
She presents an example for air pollution epidemiology. Specifically, looking at effects of installing filters on smokestacks to reduce air pollution emissions.
April 9, 2025 at 9:54 AM
She presents an example for air pollution epidemiology. Specifically, looking at effects of installing filters on smokestacks to reduce air pollution emissions.
An interesting talk by Fabrizia Mealli kicks off the #EuroCIM2025 "Interference and continuous exposures” session. I'll actually start with the conclusions slide to give you an idea of the content. 🧵...
April 9, 2025 at 9:54 AM
An interesting talk by Fabrizia Mealli kicks off the #EuroCIM2025 "Interference and continuous exposures” session. I'll actually start with the conclusions slide to give you an idea of the content. 🧵...
Nice to see some familiar faces from the Berlin causal inference community (@HertieSchool) involved in that work!
April 9, 2025 at 8:28 AM
Nice to see some familiar faces from the Berlin causal inference community (@HertieSchool) involved in that work!