gavin stewart

🌿Zeidler et al. used time-series photographic data to quantify the response of subalpine vegetation to 🌨️❄️snow cover duration. Check out their new paper! #ResearchPaper onlinelibrary.wiley.com/doi/full/10.11…

“The world has changed” says Chancellor on #bbclaurak. Yes it has - so why stick to outdated counterproductive austerity agenda & arbitrary fiscal rules? You can’t cut your way to growth. Taxes on extreme wealth of super rich would be far fairer & more effective than more cuts

My research is available on @ResearchGate: researchgate.net/publication/39…

The tuition fee rise will mean we can finally remove half of the asbestos from the Student Welfare Hub by 2029.

@JoachimSchork interesting that folks really worry about simpsons paradox/aggregation bias especially in rcts and add loads of model complexity to avoid it- but what little meta-science we have shows its often not a big problem. Arguably more parsimonious to explore heterogeneity depending....

@metanutter Thanks for sharing your insights, Gavin!

I recently made a very popular LinkedIn post about Simpson's Paradox, which resulted in an engaging conversation. Paul Julian made a great comment on the relationship between Mixed Effects Models and Simpson's Paradox that I wanted to share with you. He pointed out that when specified correctly, Mixed Effects Models can avoid being fooled by Simpson's Paradox. Unlike a naive linear model that analyzes all data at once, which might lead to misleading conclusions, mixed effects models separate fixed effects (consistent effects across all groups) and random effects (group-specific deviations from the overall trend). This allows the model to account for variations both within and between groups, leading to more accurate interpretations. In the plot below (generated from reproducible code – thanks, Paul!), you can see how different models compare: 🔹 Fixed Effect (black line): Captures the overall relationship, assuming it is the same across all groups. 🔹 Group Linear Model (dashed red line): Shows the trend within each subgroup, revealing how group-specific relationships can differ. 🔹 Naive Linear Model (gray line): Fails to account for subgroup differences, which can lead to misleading conclusions due to Simpson's Paradox. 🔹 Random Effect (blue line): Captures the variation between groups, allowing for group-specific deviations from the fixed effect. Here's the original post: linkedin.com/posts/joachim-… Important Notes: Mixed effects models offer a flexible framework to address Simpson's Paradox, effectively capturing both group-level and overall trends. However, they have limitations and alternative approaches should be considered. Mixed models, like any statistical tool, can be mis-specified if key variables are omitted. In certain cases, simpler models like OLS can handle group effects just as effectively, provided the predictors are correctly specified. For longitudinal or clustered data, marginal models like GEE or MMRM may be better suited when the goal is to estimate population-average effects, especially since mixed models focus on conditional, subject-specific effects. Additionally, Simpson’s Paradox requires careful causal understanding. Grouping variables can either be confounders or colliders, which influences the choice of model. An inappropriate adjustment can lead to incorrect conclusions, making it crucial to understand the causal structure before deciding whether to use a mixed model or a simpler approach. For regular tips on data science, statistics, Python, and R programming, check out my free email newsletter. More information: eepurl.com/gH6myT #R4DS #RStats #DataAnalytics #Statistical #database #DataVisualization

@Watts_SH @StirUni @StirlingScience @ForestryTrust @WoodlandTrust @WTScotsocial @N_T_S @MacaulayDvTrust @Forest_Research looking forward to reading this...important work...

📢Delighted to announce my big new #publication in the journal #Restoration #Ecology is now available online & #OpenAccess: "Montane willow scrub restoration in Scotland: reviewing 30 years of progress to reestablish the altitudinal treeline" (🧵1/10) 👉onlinelibrary.wiley.com/doi/10.1111/re…

You know how important you are by how many offices you have. I’ll spend the morning in my faculty office and the afternoon in the dean’s suite. So, let me know when you can leave your carrel to come meet and I’ll send the location.

@JMBecologist its not hard to pre-register- methods and priors. It could clean up hacking and harking and still let ecologists explore

Ecology would be richer if researchers were able to say “I wasn’t expecting that!” when a study has unexpected outcomes Rather than the post hoc hypothesis setting which is obvious - yet not admitted - in some papers So much of ecology remains exploratory - let’s embrace that

As another snow-spotting season draws to a close, it’s worth mentioning an important point. All the trips we do, all the gear we buy, all the fuel we need, all the sandwiches we buy to keep us going, all the sacrifices of time and effort, is done entirely at our own expense. Though we appear regularly in the media, and the results of our work are becoming well known, (and cited) only one educational or governmental institution* in Scotland has ever shown the slightest bit of interest in collaborating with us. Why is this, I wonder. Lack of funds? Lack of interest? Despite what many people think, I don’t do this for a living. It’s very much an amateur interest. In fact, none of us do it for a living. We do it for the pure love of it. This isn’t a moan, incidentally, it’s just an observation. * The National Museum of Scotland is an honourable exception. They have an exhibit dedicated to it.

#SORTEE2024 had: 🗺️ 268 people registered from 35 countries 💬 2 plenaries, 3 social mixers, & 15 interactive events 🙌 37 facilitators & 11 volunteers Help us next year! tinyurl.com/volunteerSORTE… Recordings feat. @hcp4715, @KateLaskowski, & @rose_odea: tinyurl.com/SORTEE2024plen…

Such a relief that there’s no climate crisis so these developments can go ahead without concern & without undermining commitments to being the greenest government ….

Congratulations to Nihon Hidankyo, a group of survivors of the atomic bombs dropped on Hiroshima and Nagasaki, for winning the Nobel Peace Prize. At a time when the threat of nuclear weapons is growing – and the taboo against threatening their use is weakening – Nihon Hidankyo’s work reminds us that these weapons have a terrible human cost, and that the pursuit of a world without nuclear weapons is an essential part of making sure we leave our children a safer, more secure world. Nihon Hidankyo built a powerful movement out of personal tragedy and insists upon our common humanity.

Just Fancy That ! "Vice chancellors' huge pay packages reflect responsibilities of running large organisations", said @UniversitiesUK "Financial position of #universities has deteriorated quickly in past year, some are going bankrupt ", said UniversitiesUK @PlashingVole

It’s as if none of the work on sustainable economies ever happened. Starmer bangs on about “growth” but growth of what, for whom, at what cost? Not all growth is good & as scientists concluded last week “In a world of finite resources, unlimited growth is a perilous illusion”.

94 yrs ago, Charles Elton asserted: "The 'balance of nature' does not exist and perhaps never has existed. The numbers of wild animals are constantly varying to a greater or less extent, and the variations are usually irregular in period and always irregular in amplitude." 🧵1/3

@DiaryofaSickGrl bad science

What’s an illness that needs more research?

@SolomonKurz depends what you want to know and what field you're working in. Generally speaking- really good guidance from both Cochrane and Campbell. Research Synthesis methods is the go to journal...

What are the best textbooks on meta-analysis?