Video map

Every video here comes from some chapter’s "watch next" section, with a link back to where it was recommended. The list is assembled from the pages rather than kept by hand, so it cannot drift from what the chapters actually suggest.

Chinese-language videos

繁中臺大開放式課程 NTU OCW· 92 minA National Taiwan University lecture, spoken in Mandarin with English slides. The most complete derivation of the design prototypes available free.

From How to read a clinical research paper

繁中臺大開放式課程 NTU OCW· 54 minA full National Taiwan University lecture in Mandarin. The first two sections of this chapter are derived far more completely there.

From Case-control study

繁中臺大開放式課程 NTU OCW· 68 minMandarin, same course. Mantel-Haenszel methods and selection bias, which is exactly what sections four and five here are about.

From Case-control study

繁中Cochrane Taiwan· 96 minIn Traditional Chinese. A hands-on demonstration of turning a PICO question into a search string — the practical companion to section three.

From Systematic review and meta-analysis

繁中杜裕康老師研究室· 11 minIn Traditional Chinese. On choosing a question that can actually be answered — get this step wrong and no amount of elegant statistics later will rescue it.

From Systematic review and meta-analysis

繁中Colon & Code· 16 minThe only Traditional Chinese video that tackles the p-value question head on and argues for the SMD instead. It covers exactly what sections three and four of this page do.

From Table 1 and standardised mean differences

繁中Colon & Code· 5 minA five-minute condensed version covering the p-value trap, normality testing and missing values in one go. Good to watch before reading this page.

From Table 1 and standardised mean differences

繁中EDMAN MURMURS· 10 minTraditional Chinese, written for clinicians. Ten minutes covering the t-test alongside its non-parametric alternatives.

From t-tests and analysis of variance

繁中EDMAN MURMURS· 14 minThe sequel to the previous video — what ANOVA compares, and why it is not the same as running many t-tests.

From t-tests and analysis of variance

繁中EDMAN MURMURS· 14 minIf the difference between SD and SE is not solid, the confidence interval and central limit theorem sections here will be hard going. Watch this first.

From t-tests and analysis of variance

繁中臺大開放式課程 NTU OCW· 4 minA four-and-a-half minute unit from National Taiwan University's open courseware on type I and type II error — the accumulating type I error is exactly what the multiple comparisons section uses.

From t-tests and analysis of variance

繁中EDMAN MURMURS· 10 minTraditional Chinese and clinically oriented. Ten minutes on comparing observed with expected counts — worth watching before this page.

From Chi-square test and Fisher's exact test

繁中EDMAN MURMURS· 11 minOnce a chi-square test is significant, the effect measure you report is an RR or an OR. This video is entirely about the difference and which design each belongs to.

From Chi-square test and Fisher's exact test

繁中EDMAN MURMURS· 9 minMost wrong test choices start with a misread variable type. Watch this when you are unsure whether a variable is nominal or ordinal.

From Chi-square test and Fisher's exact test

繁中EDMAN MURMURS· 13 minIn Mandarin, from a clinician's point of view. Useful if you need the Chinese terminology for coefficients and R² alongside the English.

From Linear regression

繁中臺大開放式課程 NTU OCW· 22 minIn Mandarin. Lists the assumptions of simple linear regression one by one — the same list the sixth section of this page is checking.

From Linear regression

繁中NYCU OCW· 143 minIn Mandarin. Two hours of full derivation. Watch it if you want the algebra behind the formulas; skip it if your goal is reading papers.

From Linear regression

繁中EDMAN MURMURS· 13 minIn Mandarin, from a clinician's point of view. A full episode, useful if you need the Chinese terms alongside the English ones.

From Logistic regression and the odds ratio

繁中MeDA School（洪弘）· 45 minIn Mandarin, graduate-level hands-on R. Watch it if you want to type the whole workflow out yourself.

From Logistic regression and the odds ratio

繁中MeDA School（洪弘）· 41 minA graduate-level lecture on generalised linear models, in Mandarin. No verified Chinese-language video covers count regression on its own; this is the closest.

From Poisson and negative binomial regression

繁中Ming-Chieh Shih· 16 minIn Mandarin, starting from the data structure. Useful if you need the Chinese terminology alongside the English.

From Censoring and the structure of survival data

繁中MeDA（臺大公衛洪弘教授）· 44 minA full graduate-level lecture in Mandarin. Watch this one if you want the mathematics underneath censoring.

From Censoring and the structure of survival data

繁中EDMAN MURMURS· 11 minIn Mandarin, aimed at clinicians. It is about what to read off the figure rather than how to compute it.

From Kaplan-Meier curves and the log-rank test

繁中腫瘤科吳教恩醫師· 9 minWalks through a real oncology trial end to end. Best watched after you have finished this page.

From Kaplan-Meier curves and the log-rank test

繁中Ming-Chieh Shih· 18 minIn Mandarin, continuing the first part recommended on B3-01. This episode covers the Cox model and the hazard ratio.

From The Cox proportional hazards model

繁中MeDA（臺大公衛洪弘教授）· 51 minIn Mandarin, from National Taiwan University's public health programme — the only full lecture in the site's inventory that covers assumption diagnostics and residuals. Go here for the mathematics.

From The proportional hazards assumption and Schoenfeld residuals

繁中MeDA（臺大公衛洪弘教授）· 19 minA hands-on R session in Mandarin showing how survival data is prepared and modelled. Watch it first if you want the code on this page to actually run.

From Time-dependent covariates

繁中EDMAN MURMURS· 9 minIn Mandarin, from a practising clinician's angle. Useful if you also need the Chinese terms these concepts go by in Taiwanese teaching hospitals.

From The 2x2 table, sensitivity and specificity

繁中臺大開放式課程 NTU OCW· 60 minA full lecture in Mandarin from National Taiwan University's open courseware. The verification bias in this page's last section gets its wider epidemiological context here.

From The 2x2 table, sensitivity and specificity

繁中rookie days 菜鳥日子· 31 minIn Mandarin, and clearer than most textbooks on how screening differs from diagnosis — the subject of this page's fifth section.

From Predictive values and prevalence

繁中獸醫好想告訴你· 19 minThe only full episode in Mandarin that takes the positive likelihood ratio head-on, and the place to find the Chinese terms.

From Likelihood ratios and the Fagan nomogram

繁中EDMAN MURMURS· 13 minFirst episode of a six-part series in Traditional Chinese; useful if you need the Chinese wording for these terms.

From ROC curves and the area under them

繁中EDMAN MURMURS· 13 minIn Traditional Chinese, and it lands exactly on the two blind spots covered in sections five and six here.

From ROC curves and the area under them

繁中EDMAN MURMURS· 16 minIn Traditional Chinese, and it maps onto sections two and four here — why "optimal" depends on costs.

From Choosing a cut-off, and comparing two ROC curves

繁中EDMAN MURMURS· 17 minIn Traditional Chinese, on the optimism bias of picking a cut-off in the same data; its treatment complements this one.

From Choosing a cut-off, and comparing two ROC curves

繁中Colon & Code· 10 minIn Mandarin. The first half is exactly the point this page opens with — a prediction question and a causal question are not the same question.

From Variable selection for prediction models

簡中Bessie Hiram· 13 minIn Simplified Chinese. Walks through the mechanics of internal validation end to end, and is the easiest place to pick up the Chinese equivalents of the terms on this page.

From Internal validation and optimism

簡中Bessie Hiram· 17 minIn Simplified Chinese; walks through a calibration plot from drawing it to reading it, and gives the Chinese terminology alongside the English.

From Calibration

繁中Colon & Code· 6 minThe only short Traditional-Chinese video that tackles DAGs head on; it maps onto the third section here.

From Confounding, DAGs and what to adjust for

繁中臺大開放式課程 NTU OCW· 63 minA complete lecture in Traditional Chinese from National Taiwan University's open courseware. If you want causal inference taught as a course, this is the textbook version of this page.

From Confounding, DAGs and what to adjust for

繁中資料科科講· 4 minThe shortest Traditional-Chinese introduction; four minutes to build a mental picture before reading this page.

From Propensity score matching

繁中陳冠甫（長庚）· 64 minA full lecture in Traditional Chinese; every section of this page has a longer version in it.

From Propensity score matching

繁中李昱老師· 179 minThe only Traditional Chinese course on instrumental variables, and it is econometrics — there is no clinical version in Chinese. Worth it if you want the mathematics in your first language, but every example is an economics one.

From Instrumental variables

繁中Sardinosis· 10 minA Traditional Chinese introduction written from a medical student's point of view. Watch it to get the whole map of "what is being pooled" before working through the details on this page.

From Effect measures and their variances

繁中EDMAN MURMURS· 9 minIn Traditional Chinese, and built around the question "why weight at all" — which is precisely what the third section here takes apart.

From Fixed-effect and random-effects models

繁中杜裕康老師研究室· 11 minThe clearest Traditional Chinese treatment of heterogeneity there is, taught by a biostatistics professor at National Taiwan University. Worth watching before the first two sections here.

From Heterogeneity — I², τ² and prediction intervals

繁中杜裕康老師研究室· 10 minPicks up where the first video leaves off and ties heterogeneity to small-study effects, which leads straight into the last section here and into B7-04.

From Heterogeneity — I², τ² and prediction intervals

繁中杜裕康老師研究室· 15 minTransitivity is the foundation the whole method stands on, and it is not something a test delivers. Worth watching before the first section here.

From Network meta-analysis — indirect comparison and treatment ranking

繁中杜裕康老師研究室· 17 minWorks through one analysis from the shape of the data to the output. This page uses R, but what you look for in the output is identical.

From Network meta-analysis — indirect comparison and treatment ranking

繁中杜裕康老師研究室· 18 minThe most complete treatment of ranking metrics available in Traditional Chinese, matching the last section here.

From Network meta-analysis — indirect comparison and treatment ranking

繁中杜裕康老師研究室· 19 minThe extension for when a "treatment" is really several components (fixed-dose combinations, add-on therapy). Not covered here, but worth knowing the route exists.

From Network meta-analysis — indirect comparison and treatment ranking

English-language videos

ENThis Is Why with Dr. Busti· 51 minIf you want the whole map of study designs before anything else, this covers them in one sitting. Skip it when time is short — section two here is the condensed version.

From How to read a clinical research paper

ENTerry Shaneyfelt· 6 minFive and a half minutes of overview. Watch it first and the opening section here is easier to picture.

From Cohort study

ENMichael Fralick· 6 minThe cleanest short account of where cohort and case-control designs part ways.

From Cohort study

ENTerry Shaneyfelt· 5 minFive minutes of overview. Get your bearings here before reading the chapter.

From Case-control study

ENHenrik's Lab· 3 minThree and a half minutes, mostly diagrams.

From Case-control study

ENHenrik's Lab· 3 minThree and a half minutes on what randomisation is actually solving. Worth watching before this chapter.

From Randomised controlled trial

ENTerry Shaneyfelt· 5 minSpecifically on why you cannot drop a patient who did not take the drug. Pairs with step four.

From Randomised controlled trial

ENCochrane Austria· 7 minGoes further than the previous one — ITT against per-protocol and as-treated, with worked cases.

From Randomised controlled trial

ENThe Pharmacist Academy· 3 minThe model in this chapter is a Cox model. Make sure your reading of the HR is solid first.

From Prediction model study

ENBiostatsquid· 11 minThe bridge from regression to risk prediction. Watch it before the "Development" section below.

From Prediction model study

ENResearch Masterminds· 9 minSeparate the three kinds of review before anything else — that is exactly what section one below does.

From Systematic review and meta-analysis

ENDr Amina Yonis· 18 minWalks the whole process from PICO to PRISMA. Worth watching before you actually start one.

From Systematic review and meta-analysis

ENTerry Shaneyfelt· 8 minAppraising a systematic review as a reader. Read it against the "common misuses" table at the end of this chapter.

From Systematic review and meta-analysis

ENBedroom Medicine· 5 minWatch this if you want to see the package produce the whole table in one line — the part this page deliberately does by hand.

From Table 1 and standardised mean differences

ENRverse Analytics· 1 minA one-minute demonstration of gtsummary, the other common choice alongside tableone.

From Table 1 and standardised mean differences

ENStatQuest with Josh Starmer· 9 minNine minutes on what least squares is actually doing. Watch it before the third section and "minimise the sum of squared residuals" stops being a slogan.

From Linear regression

ENStatQuest with Josh Starmer· 9 minNine minutes to build the intuition — why probability gets converted to log-odds and where the S-shaped curve comes from. Watch before the first section.

From Logistic regression and the odds ratio

ENStatQuest with Josh Starmer· 19 minSpecifically on how a coefficient becomes an OR, which is exactly the column the fourth and fifth sections here are reading.

From Logistic regression and the odds ratio

ENVery Normal· 12 minPuts linear, logistic and Poisson regression inside one framework. Watch it and the three chapters turn out to be the same thing with a different link function.

From Poisson and negative binomial regression

ENzedstatistics· 14 minThe best companion to this page. It separates censoring from truncation more clearly than most textbooks; watch it after the second section here.

From Censoring and the structure of survival data

ENnumiqo· 13 minA thirteen-minute overview for anyone completely new to survival analysis — get the map first, then come back for the details.

From Censoring and the structure of survival data

ENnumiqo· 10 minTen minutes drawing the KM calculation out step by step. The formula below reads much more easily afterwards.

From Kaplan-Meier curves and the log-rank test

ENnumiqo· 6 minA six-minute tour of the model, with animation that makes "the hazard ratio stays constant" visible.

From The Cox proportional hazards model

ENTileStats· 14 minTakes the word "proportional" apart one picture at a time. The best visual companion to this page's first section.

From The proportional hazards assumption and Schoenfeld residuals

ENMarinStatsLectures· 14 minDerives the model and shows where the assumption enters. Watch it to see that this is not an add-on check but part of the model's definition.

From The proportional hazards assumption and Schoenfeld residuals

ENEpidemiology Stuff· 10 minAn epidemiologist's framing, focused on how conclusions bend when the assumption fails. Pairs with the last two sections here.

From The proportional hazards assumption and Schoenfeld residuals

ENNienke de Glas, MD PhD· 9 minEight minutes from a clinical epidemiology angle — the cheapest way into this topic. Watch it before the second section below.

From Competing risks — cumulative incidence and Fine-Gray

ENAmerican Joint Replacement Research-Collaborative· 8 minUses a joint replacement registry — "revision surgery versus the patient dying first" is structurally identical to the MGUS example on this page.

From Competing risks — cumulative incidence and Fine-Gray

ENSAS Users· 9 minSoftware-oriented, for readers working in SAS. The concepts match the R workflow shown here exactly.

From Competing risks — cumulative incidence and Fine-Gray

ENVery Normal· 15 minIn English, rebuilding survival analysis from the definition of the hazard function. The claim on this page that "the risk set is reassembled at every event time" lands much better after watching it.

From Time-dependent covariates

ENGlobal Health with Greg Martin· 3 minThree minutes to build the mental picture of the 2×2 table. Watch it before section two and the definitions will feel inevitable rather than arbitrary.

From The 2x2 table, sensitivity and specificity

ENClinical Information Sciences· 2 minUnder two minutes, turning the concept into an arithmetic you can perform. It fills the gap the previous video leaves — that one explains but never computes.

From The 2x2 table, sensitivity and specificity

ENStatQuest with Josh Starmer· 12 minThe same definitions told once more from a machine learning angle, so that recall and precision in that literature will not throw you later.

From The 2x2 table, sensitivity and specificity

ENDirty Medicine· 11 minAll four measures in one pass, with the weight put on how PPV and NPV are dragged around by prevalence — which is exactly this page's third section.

From Predictive values and prevalence

ENMedCram· 12 minTies the four measures together inside a clinical scenario. Good for rebuilding intuition after working through the algebra in section two.

From Predictive values and prevalence

ENTerry Shaneyfelt· 10 minTen minutes covering both the definition and the bedside use. Watching this before section two is the cheapest way into the page.

From Likelihood ratios and the Fagan nomogram

ENPhysiotutors· 8 minMoves from SpPIN / SnNOUT to likelihood ratios, filling exactly the gap left at the end of the B4-01 page.

From Likelihood ratios and the Fagan nomogram

ENJAMA Network· 19 minFrom JAMA's Rational Clinical Examination series, demonstrating how likelihood ratios are actually used in real bedside decisions.

From Likelihood ratios and the Fagan nomogram

ENRahul Patwari· 8 minHand-drawn throughout, and the section on the nomogram is clearer than most slide decks manage.

From Likelihood ratios and the Fagan nomogram

ENStatQuest with Josh Starmer· 16 minAnimates the act of sliding the cut-off. Watch it before the first section and the way the curve is generated becomes concrete.

From ROC curves and the area under them

ENTerry Shaneyfelt· 5 minA five-minute clinical overview, good for returning to the diagnostic setting after StatQuest.

From ROC curves and the area under them

ENData School· 14 minFills in the probabilistic reading of AUC and how it relates to choosing a threshold — the material of the fourth section here.

From ROC curves and the area under them

ENRahul Patwari· 12 minWalks very slowly through what happens as the cut-off moves along the curve. Worth rewatching before the Youden section.

From Choosing a cut-off, and comparing two ROC curves

ENATP· 7 minA seven-minute refresher on the curve and the area under it — a warm-up for the three sections here that compare two curves.

From Choosing a cut-off, and comparing two ROC curves

ENTerry Shaneyfelt· 10 minEstablishes what a prediction rule looks like in clinical use before we argue about which variables belong inside one.

From Variable selection for prediction models

ENRichard_D_Riley· 57 minThe full hour, with variable selection as one segment of it. Worth watching once you want to see the whole B5 family as a single workflow.

From Variable selection for prediction models

ENEhsan Karim· 30 minThe middle section covers overfitting and what it does to calibration, which is exactly the first two sections of this page.

From Shrinkage and penalised regression

ENNIHR Maudsley BRC· 62 minAn hour-long tour of the whole modelling process; it makes it easier to see where penalised regression sits in the pipeline.

From Shrinkage and penalised regression

ENRichard_D_Riley· 17 minThe author of these criteria explaining them himself in seventeen minutes; it maps directly onto the third and fourth sections of this page.

From Events per variable, and the Riley sample size criteria

ENRoyalStatSoc· 71 minThe full seminar. Watch it when you want to see how the simulations behind those criteria were actually done.

From Events per variable, and the Riley sample size criteria

ENNienke de Glas, MD PhD· 7 minCovers both faces of validation in about six and a half minutes. Watch it before the section on the three things a validation study must report.

From External validation

ENData & Science with Glen Wright Colopy· 68 minThe hour-long version, given by Karel Moons. The clearest treatment anywhere of the difference between temporal, geographic and domain validation.

From External validation

ENACER Consulting· 4 minFour minutes covering what confounding is. Watch it before the second section of this page.

From Confounding, DAGs and what to adjust for

ENLeslie Myint· 6 minA six-minute introduction to the graphical language. Watch the short one first and then decide whether you want the long lecture.

From Confounding, DAGs and what to adjust for

ENEpidemiology Stuff· 21 minFills in the backdoor criterion and colliders — the full-length version of sections five through eight of this page.

From Confounding, DAGs and what to adjust for

ENMichael Fralick· 7 minAn overview in a clinical context, clear on which part of randomisation a propensity score is imitating.

From Propensity score matching

ENNEJM Group· 5 minNEJM's own animated version — good for explaining the idea to a classmate.

From Propensity score matching

ENBen Lambert· 9 minFrom an econometrics background, and it pushes the argument for why one score suffices to balance all the covariates further than the clinical channels do.

From Propensity score matching

ENBrady Neal· 11 minThe clearest walk from matching to weighting anywhere, but it is one lecture inside a causal inference course — read B6-02 first.

From Inverse probability of treatment weighting

ENLeslie Myint· 10 minStarts straight from the definition of the weights, which lines up with the second and third sections here.

From Inverse probability of treatment weighting

ENNienke de Glas, MD PhD· 6 minSix minutes, framed in clinical epidemiology — the closest thing on this list to a medical audience.

From Instrumental variables

ENMod•U (Duke)· 4 minFour minutes on why taking the long way round is what buys you a causal answer. Watch it before the second section here.

From Instrumental variables

ENMod•U (Duke)· 3 minThree minutes, one assumption at a time — it maps directly onto the third section of this page.

From Instrumental variables

ENMarginal Revolution University· 13 minEconometrics framing, but the most complete of the four; the two-stage least squares derivation is here.

From Instrumental variables

ENMeta-Analysis Academy· 4 minThe shortest possible route in. Good for anyone who has never seen a meta-analysis at all, before reading the first section.

From Effect measures and their variances

ENMichael Fralick· 43 minA full introduction given by an internist, which puts the choice of effect measure back inside the clinical question. Complements the second section here.

From Effect measures and their variances

ENCochrane Mental Health· 29 minCochrane's own teaching version. The passage on choosing a model is more practical than most textbooks manage.

From Fixed-effect and random-effects models

ENLearn Meta-Analysis· 212 minA three-and-a-half-hour workshop. Watch this one if you want to run the whole thing yourself and see what each option looks like inside the software.

From Fixed-effect and random-effects models

ENTerry Shaneyfelt· 9 minThe clinician's angle, separating clinical, methodological and statistical heterogeneity. This page only deals with the third kind, but the decision to pool at all rests on the first two.

From Heterogeneity — I², τ² and prediction intervals

ENTerry Shaneyfelt· 6 minFive and a half minutes mapping every element of a forest plot to what it means. The cheapest possible preparation for this page.

From Forest and funnel plots

ENCochrane Austria· 10 minTen minutes separating confounding, chance and bias from one another. The first section of this page — that selection bias is not confounding — gets a slower version here.

From Selection bias