iiinova — where your digital self keeps evolving

Dan Major-Smith @djsmith90.bsky.social · 28/04/2026

Hope to have a full paper in the next few months - But for those lucky enough to attend, there will be a sneak peek of preliminary results at the CES conference in a couple of weeks. W/ @bgpurzycki.bsky.social @rcc-au.bsky.social, plus too many other awesome folks to mention!

020

Dan Major-Smith @djsmith90.bsky.social · 28/04/2026

Featuring: - Individual-level cross-cultural data from over 20 non-Western societies - An explicitly causal approach with DAGs and a clear estimand - Fully-varying multi-level Bayesian "Mundlak" models - Ethnographically informed data collection Mmmmm. Deeeelicious 😋

130

Dan Major-Smith @djsmith90.bsky.social · 28/04/2026

Do supernatural punishment beliefs promote cooperation beyond parochial boundaries? And is this effect stronger in "moralistic" religious traditions (eg Christianity, Hinduism)? Working with some of best & brightest in our field, that's what our new prereg aims to answer: github.com/bgpurzycki/G...

github.com

192

Reposted by Dan Major-Smith

AnthroTools: A Package in R @anthrotools.bsky.social · 21/10/2025

"Modeling Uncertainty around Free-list Cultural Salience Scores" by @rcc-au.bsky.social's @djsmith90.bsky.social and @bgpurzycki.bsky.social OnlineFirst up now at Field Methods journals.sagepub.com/doi/full/10.... #AnthroTools #quantethnography

043

Reposted by Dan Major-Smith

Alba Fernández-Sanlés (Dr, she/her) @albafsanles.bsky.social · 23/06/2025

I am devastated because our cat Trasno is critically ill and needs urgent treatment to survive, but it’s financially overwhelming. If we’ve ever made you smile or feel supported, please consider sharing or contributing: gofundme.com/f/help-save-... Thank you from the heart, Alba

gofundme.com

Donate to Help save our cat Trasno: Urgent treatment needed, organized by Javier Ortiz De Guinea

Help save Trasno: Urgent Treatment needed for our cat’s life… Javier Ortiz De Guinea needs your support for Help save our cat Trasno: Urgent treatment needed

462

Dan Major-Smith @djsmith90.bsky.social · 10/06/2025

Our experience at @pci-regreports.bsky.social was great - Really helpful and collaborative reviews and editors, made for a refreshing change from the usual grind of the academic review/publishing process! If any folks are PCI-curious, would totally recommend! 😊

030

Dan Major-Smith @djsmith90.bsky.social · 10/06/2025

Caveat 2) We measured climate *concern*, not climate *anxiety*. It's possible that reactions to the climate crisis which are more extreme than mere 'concern' have an impact on mental health - We simply can't infer anything about this from our paper

110

Dan Major-Smith @djsmith90.bsky.social · 10/06/2025

Many caveats though (ofc!) 1) Just because we *tried* to estimate a causal effect, doesn't mean we *succeed*. Our causal estimand was our target, but our aim may have been terrible, but we at least tried to be open about these issues (unmeasured confounding, selection bias, measurement error, etc.)

100

Dan Major-Smith @djsmith90.bsky.social · 10/06/2025

What did we find? Pretty much nada! In this UK population of young adults (~30 years old) there was essentially no relationship between climate concern and subsequent mental health/well-being

100

Dan Major-Smith @djsmith90.bsky.social · 10/06/2025

Using the wonderful longitudinal ALSPAC data, we tried to answer whether concern over climate change causes subsequent mental health. Lots of previous work has looked at climate anxiety/concern and mental health, but most is X-sectional, so is almost impossible to learn anything about causality

110

Dan Major-Smith @djsmith90.bsky.social · 10/06/2025

The world may be a burning shit-storm, but we're trying to do our little bit to make science slightly better - Our new Registered Report on climate concern and mental health was just recommended/accepted at @pci-regreports.bsky.social doi.org/10.24072/pci...

doi.org

Climate change concern doesn't impact the subsequent mental heal...

131

Reposted by Dan Major-Smith

Abbey Page @abbeyepage.bsky.social · 14/05/2025

📣 New paper by myself @djsmith90.bsky.social and colleagues published today in @royalsociety.org Proc B 📣 Grandmothers are often seen as key carers, but this isn’t always the case. We explore why in Agta forgers reflecting on the implications of demography for evolution of grandmothering 👵🌍

royalsocietypublishing.org

Demography of grandmothering: a case study in Agta foragers | Proceedings of the Royal Society B: Biological Sciences

Grandmothers are often presented as key carers due to low costs and high inclusive fitness returns. Empirically, however, grandmothers are not consistently important. Understanding the factors that pr...

12410

Reposted by Dan Major-Smith

AnthroTools: A Package in R @anthrotools.bsky.social · 30/04/2025

AnthroTools 2.0 is live! Nine new functions that propagate, plot, and summarize uncertainty around item and cultural salience values. @djsmith90.bsky.social @bgpurzycki.bsky.social @rcc-au.bsky.social

053

Dan Major-Smith @djsmith90.bsky.social · 11/04/2025

Can religious beliefs shape peoples' attitudes and behaviours regarding climate change? We explored this using data from a large-scale UK cohort study (ALSPAC) In short: Perhaps, but it's complicated. Answer depends on generation, religiosity measure and climate question doi.org/10.1371/jour...

doi.org

Associations between religiosity and climate change beliefs and behaviours in the Avon Longitudinal Study of Parents and Children (ALSPAC)

Individual actions are crucial to mitigating the impact of anthropogenic climate change. Understanding the factors shaping individuals’ climate beliefs and behaviours is therefore essential to help en...

072

Dan Major-Smith @djsmith90.bsky.social · 09/04/2025

As a side note, even if you don't give a hoot about free-list data, check out the paper for the sweet use of LaTeX - Looks pretty legit, right? (What are we paying journals the big bucks for again...?)

020

Dan Major-Smith @djsmith90.bsky.social · 09/04/2025

We focus on salience scores, starting simple with bootstrapping methods, followed by more principled/bad-ass Bayesian ZOIB and ordered Beta models (they sound scary, but are super-cool). This approach can be also applied to other free-list metrics as well (Jaccard's similarity, cultural FST, etc).

110

Dan Major-Smith @djsmith90.bsky.social · 09/04/2025

Are you into free-list methods and want to incorporate uncertainty into your group-level estimates (eg Smith's S)? Of course you do! Then @bgpurzycki.bsky.social and I have a new preprint for your delectation! doi.org/10.31219/osf...

doi.org

OSF

140

Reposted by Dan Major-Smith

Aarhus University's Religion, Cognition, and Culture Unit @rcc-au.bsky.social · 19/03/2025

No clear co-causal evidence of religion and mental health! Hot off the digital press from RCC's @djsmith90.bsky.social! journals.plos.org/plosone/arti...

042

Reposted by Dan Major-Smith

Julia M. Rohrer @dingdingpeng.the100.ci · 17/02/2025

New blog post! Sometimes, when reviewing a manuscript, it's really unclear to me what precisely the authors are trying to do -- which makes it hard to evaluate the work properly. So, here's some advice for how to ensure that readers don't get lost. www.the100.ci/2025/02/17/r...

the100.ci

Reviewer notes: Avoid any ambiguity about analysis aims

For any central statistical analysis that you report in your manuscript, it should be absolutely clear for readers why the analysis is being conducted in the first place – that is, the analysis goal s...

611138

Reposted by Dan Major-Smith

Abbey Page @abbeyepage.bsky.social · 21/01/2025

1/ 🧵 New preprint out today 🤩 With @djsmith90.bsky.social and co we have done deep-dive into grandmothering in the Agta to unpick the role of demography in shaping the role of grandmothers in direct childcare 👇🧪🏺 #Grandmothering #demography #BioAnth #EvPsych

osf.io

OSF

13517

Reposted by Dan Major-Smith

Centre for Culture and Evolution @brunelcce.bsky.social · 05/12/2024

📢 The final CCE seminar of 2024 is coming up soon with @djsmith90.bsky.social leading us into the 🎄 festive 🎄season with a presentation on *Religion and mental health: a causal Christmas carol* 🎅 📆 18th Dec 2024 - 12:45 in person start with food, 13:00 online talk start 🔓Open to all

1138

Dan Major-Smith @djsmith90.bsky.social · 16/11/2024

Did not expect my first inclusion in a starter pack to be "people named Dan Smith with PhDs"! 😅 Still, at least you now know where to find us!

050

Reposted by Dan Major-Smith

Pat Savage @patrickesavage.bsky.social · 08/10/2024

I found @djsmith90.bsky.social's figure about ethical publishing so helpful for students I added it to my guide to academic reading/writing/publishing (osf.io/p37zj)

Screenshot from article including meme and figure of more and less ethical academic publishing options

021

Dan Major-Smith @djsmith90.bsky.social · 04/10/2024

Yeah, I'd reckon probably somewhere around there, or a bit to the left with the 'society' journals - technically AAAS is non-profit, and it doesn't act as much like a commercial publisher than CUP and OUP.

010

Dan Major-Smith @djsmith90.bsky.social · 04/10/2024

Ace, thanks Pat! And very cool (I mean, depressing as hell, but still very cool) - Thanks for sharing 😅

110

Dan Major-Smith @djsmith90.bsky.social · 04/10/2024

Alas, I agree with @drbeth.bsky.social that it's still super-common (eg, anecdotes about hiring decisions made on having a Nature/Science/Whatever paper)... Hopefully will start changing more soon though! 🤞 V. interested to know what you find if you do a deep-dive into this 🙂

020

Dan Major-Smith @djsmith90.bsky.social · 04/10/2024

A lot has been written about this recently, but I think it's still useful to remind ourselves just how warped and unfair the academic publishing system is The blog goes through: - How the system is unfair - How it got that way - What we can do to (try and) change it Hope it's of some use! 🙂

010

Dan Major-Smith @djsmith90.bsky.social · 04/10/2024

New blog post I cobbled together for the Bristol Reproducibility Network Ethical Academic Publishing: How to Make Academic Publishing Fairer, More Open and Less Wasteful openresearchbristol.blogs.bristol.ac.uk/2024/09/23/e...

openresearchbristol.blogs.bristol.ac.uk

Ethical Academic Publishing: How to Make Academic Publishing Fairer, More Open and Less Wasteful – Open Research at Bristol

41615

Dan Major-Smith @djsmith90.bsky.social · 03/10/2024

Tak! Excited to start working with all the wonderful @rcc-au.bsky.social folk in Aarhus too! 🙂

0100

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

As researchers we definitely still need to consider the impact of selection bias in our work, but hopefully this paper can help provide a bit more understanding about selection bias and some context for how selection might bias our results. END/

010

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

Does this mean we can all get on with our lives and safely ignore selection bias? Of course not! Different studies have different patterns and strengths of selection which could realistically result in waaaay more bias (eg, recruitment into UKBioBank: doi.org/10.1093/ije/...).

100

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

So, what does this all mean? Well, it suggests that for this specific research question, this specific amount of missing data, and these specific strengths of selection, selection bias may not substantially impact our conclusions.

100

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

Results were similar in our secondary analyses comparing full vs highly-selected ALSPAC samples, with no meaningful differences in interpretation despite large differences in sample size and risk of selection bias.

100

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

Turns out, the answer is: practically no difference whatsoever! While there was a small amount of bias – a 2%-point difference in the odds ratio (from 1.00 to 1.02) – the confidence intervals still contained the true value ~95% of the time, so for all practical purposes the bias was non-existent.

110

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

We simulated ~70% missing data, and in the strongest selection scenario religious attendance increased the odds of participation by 50% with depression lowering the odds by 20%. These are not insignificant differences, yet it’s not obvious (to me) how much impact this would have on any conclusions…

100

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

To assess the extent of selection bias, we first conducted a simulation study using a range of plausible selection parameters, followed by analyses on the actual ALSPAC data comparing the full sample to a highly-selected sample (~30%) still participating after three decades.

100

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

Using baseline ALSPAC data, we can explore the factors known to predict continued participation and see the impact of this selection on any inferences. In our study, we focused on religious attendance and depression, both of which are known to predict ALSPAC participation.

100

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

Often this is very difficult (if not impossible) to know because we simply don’t know the characteristics of people who did not participate in our study. But with some representative longitudinal datasets it is possible, such as the ALSPAC study (www.bristol.ac.uk/alspac/).

bristol.ac.uk

Avon Longitudinal Study of Parents and Children | Avon Longitudinal Study of Parents and Children | University of Bristol

100

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

Alongside confounding, measurement error and generalisability/transportability, selection bias is 1 of the 4 causal horsemen of the apocalypse we’re (rightly!) told to watch out for. But other than being told to be afraid of it, just exactly how bad is selection bias in reality?

100

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

(Still a bit confused? Don’t worry you’re in good company - Selection bias is a notoriously counter-intuitive and slippery concept. For some nice intros, see doi.org/10.1093/ije/..., doi.org/10.1038/s414... & doi.org/10.1097/EDE....)

100

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

For instance, if both our exposure and outcome predict participation (ie, collider bias), this can bias the relationship between these variables. (Image from doi.org/10.1038/s414...)

100

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

First: What the frig is selection bias? If participation in research is non-random, then our analytic sample will differ from our target population of interest. This sample-population mismatch can result in selection bias, potentially messing up all our lovely causal inferences.

100

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

We all know that – alongside global warming, Elon Musk, Vladimir Putin and other evils – selection bias is one of the largest threats to humanity* But how bad exactly is selection bias? This is what we aimed to explore in our recent paper: doi.org/10.1080/2153... *Perhaps a slight exaggeration…

296

Reposted by Dan Major-Smith

Centre for Culture and Evolution @brunelcce.bsky.social · 23/09/2024

📢📢📢💫 CCE's seminar series is kicking off in November with some amazing speakers lined up 💫📢📢📢 These are 🔓 open to all - in person and online - joining information shared closer to the date 📅 ⏲️ Talks are scheduled between 12:30 - 14:00 @Bruneluni

21915

Dan Major-Smith @djsmith90.bsky.social · 23/09/2024

Interested in causality, DAGs and human evolution? If so, then why not join the Evolutionary Human Science journal webinar on 'Causal Inference from Evolutionary Human Sciences'? Join us October 3rd 2PM BST | 3PM CEST More info here: cambridge-org.zoom.us/webinar/regi...

cambridge-org.zoom.us

UPCOMING WEBINAR: Causal Inference from Evolutionary Human Sciences Join us October 3rd 2PM BST | 3PM CEST Register Now

In recent years, researchers have made groundbreaking advances in methods for drawing strong causal inferences with observational data. These advances have also revealed that some mechanisms, with so...

034

Reposted by Dan Major-Smith

Ariel Rokem @arokem.org · 05/09/2024

New paper out! "Ten simple rules for scientific code review" journals.plos.org/ploscompbiol..., in which I detailed how to use *code review* in academic research in a didactic, effective, and enjoyable manner.

journals.plos.org

Ten simple rules for scientific code review

Author summary Scientists are increasingly writing code as part of their research. Code review is a common practice in software engineering, which entails detailed and continual examination of additio...

66038

Reposted by Dan Major-Smith

Julia M. Rohrer @dingdingpeng.the100.ci · 27/08/2024

New blog post! If you're a substantive researcher, you mainly do statistics to answer substantive questions. So let's stop putting the statistical model first. www.the100.ci/2024/08/27/l...

the100.ci

Let’s do statistics the other way around

Summer in Berlin – the perfect time and place to explore the city, take a walk in the Görli, go skinny dipping in the Spree, attend an overcrowded, overheated conference symposium on cross-lagged pane...

1214663