Quantitude

Bayes factors take center stage as two statisticians unpack what they measure, how they update odds, and how they differ from p-values. Psychic readings, Formula One rankings, replication, competing theories, default priors, and the perils of rigid evidence cutoffs bring the concepts to life. A satirical statistical psychic ad rounds things out.

Probability, gambling odds, log odds, and odds ratios take center stage, from Pascal and Fermat’s interrupted wagers to prediction markets and the bookmaker’s vigorish. The discussion explores fair bets, expected value, logistic regression, and why odds ratios are not probability ratios. Along the way: airplane hangars, Costco hot dogs, the tush push, horse steroids, and a suspicious probability of 1.8.

A circuitous, comic tour of the academic curriculum vitae, from its changing role across a career to its uses in hiring, promotion, grants, and review. Patrick and Greg explore how formatting reveals personality, how to build a master CV, and how to tailor publications, funding, teaching, presentations, and service for different audiences. Antiques Roadshow, Duolingo streaks, and a doodad drawer provide lively detours.

Paul Allison, emeritus sociologist at the University of Pennsylvania and founder of Statistical Horizons, shares his journey into quantitative methodology. He explores longitudinal fixed-effects models, within-person comparisons, change scores, difference-in-differences, and structural equation modeling. He also recalls a humbling econometrics lesson, reflects on AI’s research potential, and discusses his project on fitness pioneer Jack LaLanne.

A lively exploration of what metrics really mean, from Dulles Airport distances and oversized beers to fuel gauges, test scores, and effect sizes. The discussion challenges standardization when it hides practical context, questions labels like “small” and “large,” and makes the case for reporting multiple measures. Along the way: airport confusion, concert pitch, middle schools, and the Metric Master 3000.

A true-crime parody reopens the case against null hypothesis significance testing. The investigation traces the Fisher–Neyman–Pearson family history, examines p-value confusion, binary significance thresholds, neglected effect sizes, and the incentives behind questionable research. The verdict considers whether NHST deserves punishment, parole, or rehabilitation through better inference and scientific practice.

In this week's episode Greg and Patrick revisit their earlier promise of creating seven questions that will help you use your Quantitude to annoy the people you love. Having broken their promise, they are free to propose nine questions for which they seek a mnemonic device that does not include an acronym. Along the way they discuss relationship mandates, mildly inebriated brothers, hypoxic voice memos, fish vs. ice cubes, Smitty's corpse, talking to houseplants, statistical baggage, Henry V, Pascal's wager, My Very Educated Mother, massaging cryptids, making mischief, being cruciferous, Peter and Gary, pneumatic devices, and so that happened. Stay in contact with Quantitude!Web page: quantitudepod.orgTwitterX: @quantitudepodYouTube: @quantitudepodMerch: redbubble.com

They imagine collectible trading cards for influential philosophers of science. They riff on Popper’s falsification, Mill’s causal foundations, Kuhn’s paradigms, Cartwright’s contextuality, Lakatos’ research programs, and Meehl’s critique of weak theory. The conversation mixes serious history-of-science topics with playful banter.

They wrestle with whether college is worth it by unpacking its many purposes and who benefits. They trace college history and contrast old ideals with modern student life. They examine tangible and soft outcomes, credentialing as a signal, measurement problems, and alternative paths. They stress that value depends on individual goals, field, cost, completion, and timing.

Emily Namey, an applied qualitative researcher in public health and international development, shares her career journey and core methods. She discusses interviews, focus groups, and visual data. Short, lively segments cover using qual to inform policy and trials, coding and software, AI risks, and when qual and quant complement each other.