Ben Marwick
benmarwick.bsky.social
Ben Marwick
@benmarwick.bsky.social
Archaeologist and #rstats enthusiast
Here is my checklist summarising a small set of some of the simplest tasks you can do that have high potential to improve the reproducibility of your analysis code.

This is based on my year of reproducibility reviews for the J. of Archaeological Science:

authors.elsevier.com/a/1lHjN_6yUM... 🧪🏺
July 6, 2025 at 1:25 AM
Is archaeology a science? 🧪

Here's my new paper that has a go at answering this question by analysing 10,000 journal articles:

authors.elsevier.com/a/1lHjN_6yUM...
June 18, 2025 at 4:59 PM
Here is an excellent checklist for reproducible research, this is from the Biometrical Journal, but is relevant for computational scientists in all fields:
onlinelibrary.wiley.com/page/journal...

Lots of best practices and specific tips for increasing the quality, usefulness & impact of papers

🧪
April 2, 2025 at 7:31 PM
Our new paper reports a complete Quina technological system in the 60-50 ka assemblage at Longtan, Southwest China

Ruan, Q. et al. (2025) Quina lithic technology indicates diverse Late Pleistocene human dynamics in East Asia doi.org/10.1073/pnas...

🆓 faculty.washington.edu/bmarwick/PDF...
March 31, 2025 at 8:07 PM
Check out this impressively detailed and user-friendly documentation by @saa-aap.bsky.social on how to make computational archaeological results reproducible, thanks to Alan Faharani (contributing from American Antiquity, which has an identical policy):

www.cambridge.org/core/journal... 🧪
March 18, 2025 at 9:57 PM
I'm trying to access the Archaeology talks here underline.io/events/456/s... and being asked to pay, is there a different way to freely access the content?
February 23, 2025 at 7:53 PM
Computational reproducibility exists within a continuum of attributes, including being re-runnable, repeatable, reusable and replicable.

Making your computational results reproducible is critical for ensuring transparency, reliability and reusability 🧪
February 3, 2025 at 7:51 PM
Here are some excellent guidelines to improve the computational reproducibility of scientific 🧪 research, based on an analysis of 168 research papers.
January 29, 2025 at 6:40 AM
New paper with colleagues from Sichuan University & the Guizhou Institute of Cultural Relics and Archaeology:

"New Evidence of Human Occupation in Southwest China Since 44,800 Years ago"
December 29, 2024 at 12:54 AM
Introducing the Associate Editor of Reproducibility for Advances in Archaeological Practice doi.org/10.1017/aap....

This initiative contributes to making archaeology more transparent and accessible and a source of trustworthy and reliable information about the human past.
September 18, 2024 at 2:14 PM
Code and data are openly available online here:

zenodo.org/doi/10.5281/...
August 19, 2024 at 4:44 AM
Given this innovative approach, we placed considerable emphasis on systematically exploring the impact of both data and models on the results. This is achieved by incrementally introducing complexity, thereby laying a robust foundation for future research in this direction.
August 19, 2024 at 4:38 AM
We also used skyline analysis to examine birth, death, diversification, and turnover rates across the four major climatic warming and cooling events during this timeframe, based on the Greenland ice-core event stratigraphy.
August 19, 2024 at 4:36 AM
We used a fossilized birth-death sampling process model to infer time-scaled Bayesian phylogenies, utilizing the projectile point outline shape as continuous characters (first use of this in archaeology that we are aware of).
August 19, 2024 at 4:35 AM
New paper! Led by David Matzig, we used a state-of-the-art Bayesian phylodynamic framework to explore the evolution of projectile point shapes during the European Final Palaeolithic and earliest Mesolithic (approximately 15-11ka BP).

doi.org/10.1098/rsos...
August 19, 2024 at 4:33 AM
New paper, celebrating the work of Julie Stein on the occasion of her retirement!

DiCiro, A., Mitchell, N., & Marwick, B. (2024). Everything is a Deposit: An Interview with Pioneering Geoarchaeologist Julie K. Stein. Bulletin of the History of Archaeology, 34(1), 15. doi.org/10.5334/bha-...
June 18, 2024 at 2:07 PM
Here is our poster "Careers in Ruins: Academic Archaeology Job Trends From 2013-2023"

It was presented today at #SAA2024 #saa2024nola by
#UniversityOfWashington undergrad Anne Poole. Co-authors include Ailin Zhang, Setareh Shafizadeh & Jess Beck.

Data & #rstats code at github.com/benmarwick/a...
April 20, 2024 at 5:59 AM
The Journal of Archaeological Science has just introduced a 'Reproducibility Prize for papers that share in a transparent, clear and detailed manner their data, protocols and/or code', and announced the first winners. Congrats Andrew & Xavier!

www.sciencedirect.com/journal/jour...
November 22, 2023 at 1:55 AM
Yes, I agree it will be tricky to get right. I like the workflow here doi.org/10.1061/(ASC..., the editor there seems satisfied with it, with many papers: ascelibrary.org/jwrmd5/repro... I've got +ve reactions when presenting this to editors of archaeology journals as a viable implementation.
October 15, 2023 at 11:59 PM
I've definitely heard that concern from editors. I'm proposing they take an opt-in approach where R-using archaeologists volunteer to get their code reviewed and the reproducibility of their analysis acknowledged by the journal in the version of record, e.g. with a badge or short report to boost...
October 15, 2023 at 11:38 PM
Those 4Rs are a neat workflow. This could be perfect for several archaeology journals that publish papers with code. Just like what American Naturalist does.
October 10, 2023 at 8:39 PM
New paper! "Women in the Lab, Men in the Field? Correlations between Gender and Research Topics at Three Major Archaeology Conferences" www.tandfonline.com/eprint/S6XGC... Open access pre-print: osf.io/fsujr Code & data: doi.org/10.17605/OSF...
October 2, 2023 at 7:32 PM