Dr William Kay (he/him)

📢Extremely excited to host a #BES2025 Thematic Session next week!📈The future of quantitative learning and literacy in ecology | Thematic Session | Tues 16th Dec | 16:00-18:00 | Sidlaw Auditorium🌱 @FiebergJohn @nhcooper123 @lucaborger @TeicherCamila @chrissuthy @TiagoALOMarques

Today's the day! This is looking to be a very popular session! Exciting!

@selcukorkmaz I find the easiest way to remember the exact definition of frequentist CIs is to remember: "it's all to do with replication". If you train your brain such that when you think of CIs you immediately hear the term "replication", it helps to always get that definition right.

Avoid misconceptions about CIs A frequentist confidence interval is an interval that we calculate from data in order to make a guess at where the true value of something (like an average or effect) might be, but it isn't a statement of the probability about the true value itself. Say you're trying to estimate the average height of all adults in a town. You can't possibly measure everybody, so you take a random sample and construct a confidence interval. Say, 170 cm to 180 cm at the 95% level. That is: • If you repeated the entire procedure (take a new random sample, find the same type of interval) again and again a very large number of times, then approximately 95% of the intervals would contain the true average height of the whole population. • For your specific interval (170–180 cm), the true average either is in that range or it is not. No probability is in the picture anymore once you've seen your data. The most common error is to say, "there's a 95% chance the actual mean is between 170 and 180." That's wrong in the frequentist sense. The 95% refers to the long-run rate of success of the procedure, not the probability for this specific interval. In practice, the interval 170 cm to 180 cm tells you the interval of plausible values for the actual average height of the city, in your sample. You do not know the actual average, but the confidence interval gives you a quantitative sense of precision and uncertainty: • The interval width (10 cm) indicates how unsure your estimate is — a smaller range indicates a more accurate estimate, a larger range indicates more uncertainty. •The interval itself (170–180 cm) gives an estimate of where the true mean is expected to lie, based on your data and technique.

I'm excited to share that this Wednesday (10th Sept at 12:00 BST) I'll be presenting this year's first National Centre for Statistical Ecology @NCSE_UK invited seminar: "How should we train the next generation of statistical ecologists?" #ecology #rstats

Very wet and early start this morning to travel to Liverpool for CETL-MSOR 2025!

On the move again, to the Year 2 Coastal Biology field course in Dale, Pembrokeshire.

Thoroughly enjoyed my day return to @UofGlasgow #UKCOTS25! Another city with stunning architecture.

Interested in external training courses on biological data analysis (and more!)? Follow @pr_stats! I've taught on their courses and sent students to them. I'd highly recommend! I attended my first spatial analysis course with them back in 2016. Website: prstats.org

Learn how to apply linear and generalized linear mixed models using R prstats.org/course/introdu… #RStats #MixedModels #StatisticalModelling #EcologicalModelling #LifeSciences #OnlineCourses #DataScienceTraining #GeneralisedLinearModels #RStudio #StatisticalEcology #EcologyCourse

Online Course - Live and Online Registration: prstats.org/course/online-… #SpeciesDistributionModelling #BayesianStatistics #EcologicalModelling #RStats #QuantitativeEcology #SpatialEcology #EnvironmentalDataScience #Biostatistics #EcologyTraining #StatisticalModelling

Thanks to The Tab for covering much of our campaigns over the last few months. Just to clarify: we have ruled out industrial action *during the current IA mandate period*, which runs until the end of September 2025, not the end of the calendar year. thetab.com/2025/05/01/car…

Fantastic news for Cardiff University Meanwhile, Cardiff Metropolitan University continues to threaten staff with compulsory redundancy. @CardiffUCU @CymruUCU @DrJoGrady

This from Radio Wales is the most comprehensive take yet on @cardiffucu’s win in the @CardiffUni dispute. Worth a listen if you’re staff, a student, a trade unionist, or are just interested in the detail of what’s (& what fights remain ahead). @UCU ✊🔥💖 youtu.be/l3hef_sm6SM?si…

@BBCWalesNews with excellent, in-depth coverage of my union branch’s bittersweet industrial relations victory in defence of @CardiffUni jobs. I’m interviewed for @cardiffucu, & also some of our *amazing* students, about the need to keep fighting. ✊ @ucu youtu.be/imSu7dkCjMU?si…

Alongside herculean efforts of members & non-members -including our fab sister unions @unitetheunion & @UNISONinHE- to persuade UEB, what tipped the balance was a threat of industrial action. On #MayDay2025 we remember & reassert the power of organised labour to change lives ❤️🖤

We, the staff and students, are the University. And there has never been a better time to join a Union. ucu.org.uk/join

This is a small win and it's not the end of the story. We'll continue to build an even stronger mandate and to press upon management both the route to a better future for our University - including permanently ruling out CRs - and the strength of our Unions here at Cardiff.

We've called off planned industrial action thanks to an agreement to rule out compulsory redundancies for this year. This is not the end of the fight. I and over 1,200 of my colleagues remain "in scope", but we at least wont be served any compulsory redundancy papers this year.