iiinova — where your digital self keeps evolving

Abdullah Abdelaziz @abduaziz.bsky.social · 05/09/2026

It would be great to share your advice or recommendations, especially since so many resources out there can leave people confused, distracted, and unsure where to begin or how to climb the ladder step-by-step. (2)

100

Abdullah Abdelaziz @abduaziz.bsky.social · 05/09/2026

One suggestion, if I may. I am curious whether you have a teaching ladder or manifesto. Let's say someone who loves epi, math, and causality wants a didactic approach (books/courses). (1)

100

Abdullah Abdelaziz @abduaziz.bsky.social · 05/09/2026

I really like your content, and I'm so grateful that you make it available to everyone!

010

Reposted by Abdullah Abdelaziz

Aaron Mitchell @thewonkologist.bsky.social · 11/11/2025

Our new study is out today in @jamanetworkopen.com! We looked at trends in biologic drug prices after biosimilar competition. Key takeaways: savings from biosimilars are substantial, but take longer to accrue than commonly seen with generic small molecules. 1/ jamanetwork.com/journals/jam...

jamanetwork.com

Biologic Drug Prices in Medicare Part B After Entry of Biosimilars to the Market

This cohort study evaluates the associations of biosimilar entry to the market with the prices of originator biologics under Medicare Part B reimbursement.

143

Abdullah Abdelaziz @abduaziz.bsky.social · 03/07/2025

It's probably an unpopular opinion, but the more I do research, the more I realize that it's almost IMPOSSIBLE to do transparent and reproducible research without strong programming skills.

271

Abdullah Abdelaziz @abduaziz.bsky.social · 27/05/2025

Will the recordings be available on YouTube after the event?

110

Abdullah Abdelaziz @abduaziz.bsky.social · 11/12/2024

I am trying to access the newest version of @miguelhernan.bsky.social 's Causal Inference What If book, but it seems that Harvard's SPH website no longer shows it. Any idea how to get to the old page? #EpiSky

230

Abdullah Abdelaziz @abduaziz.bsky.social · 01/12/2024

Is there an advantage to go deep into rust compared to C++ for the R user?

110

Reposted by Abdullah Abdelaziz

Jeremy Howard @howard.fm · 25/11/2024

I've never, personally, had any direct utility from traditional academic peer review for my scientific work. I've never used that peer review as any kind of signal or credential in judging a paper or deciding whether to read it. Nearly all the papers I read are preprints.

A random snapshot of my Zotero account, showing that nearly all the papers are from arxiv

1524717

Abdullah Abdelaziz @abduaziz.bsky.social · 21/11/2024

I think it would be easier to use dput()

000

Reposted by Abdullah Abdelaziz

nddpsych @nddpsych.bsky.social · 21/11/2024

How to write an effective journal peer review using a staged writing approach: a best-practice guide for early-career researchers | International Journal of Epidemiology | 2024 #MedEd academic.oup.com/ije/article/...

academic.oup.com

How to write an effective journal peer review using a staged writing approach: a best-practice guide for early-career researchers

Abstract. Journal peer review is a gatekeeper in the scientific process, determining which papers are published in academic journals. It also supports auth

22515

Abdullah Abdelaziz @abduaziz.bsky.social · 25/09/2024

This reminds me of days navigating why I have a much smaller sample size while building a cohort for my study. It turns out that Excel removes leading zeroes in MRNs, and many patients can not be linked. Ironically, ppl out there might accuse you of complicating things when you use R

010

Abdullah Abdelaziz @abduaziz.bsky.social · 06/09/2024

For the 2nd equation to work, you assume some kind of equal confounding i.e. Somehow A <- U -> Y has the same direction and magnitude to A <- U -> W Correct?

100

Abdullah Abdelaziz @abduaziz.bsky.social · 06/09/2024

So the goal of the first equation to kind of isolate the association presented in the DAG as A <- U -> W then using it. This is why you put Z. Otherwise, your association would capture the path A <- Z -> U -> W too. Did I get it?

100

Reposted by Abdullah Abdelaziz

Andrew Heiss @andrew.heiss.phd · 04/04/2024

New blog post! Seven (7!) new tidyexplain-esque animations showing how {dplyr}'s mutate(), summarize(), group_by(), and ungroup() all work together #rstats #dataskyence www.andrewheiss.com/blog/2024/04...

Illustration showing rows in a dataframe getting grouped, then adding a new column

Illustration showing rows in a dataframe getting grouped, summarized, and ungrouped

Illustration showing rows in a dataframe getting grouped into two groups, then summarized and ungrouped

Note from the blog post showing links to download all these videos

16631

Reposted by Abdullah Abdelaziz

Jordan Nafa @ajordannafa.com · 24/12/2023

0163