Categories

All (48)

anova (4)

assumptions (1)

biostatistics (10)

categorical-data (2)

cdisc (3)

clinical (9)

confidence-intervals (1)

consulting (2)

correlation (1)

data-quality (1)

datasets (1)

dataviz (1)

diagnostics (1)

distributions (1)

effect-size (1)

energy (1)

estimation (1)

forecasting (1)

ggplot2 (3)

gtsummary (2)

hypothesis-testing (5)

markdown (1)

p-values (1)

pdf (1)

power-analysis (1)

pROC (1)

psychometrics (1)

quarto (8)

R (1)

r-tutorial (18)

reference (1)

regression (2)

reporting (4)

reproducibility (2)

sample-size (1)

simulation (1)

statistics (23)

study-design (2)

survival (1)

survival-analysis (1)

teaching (1)

time-series (1)

websites (1)

workflow (6)

Blog

Notes on R, statistics and reproducible workflows.

ADaM Derivations in R with admiral: a Mini ADSL, End to End

clinical

cdisc

biostatistics

Derive an ADSL-style analysis dataset in R with the pharmaverse admiral package: treatment variables, analysis flags, age groups and time-to-event variables — every step traceable.

Is R Accepted by the FDA? The State of R-Based Submissions

clinical

cdisc

What the R Consortium Submissions Working Group pilots actually demonstrated, why ‘validated software’ is a property of your process rather than a licence, and what a defensible R stack looks like for CROs.

Clinical TLFs in R: gtsummary and rtables Side by Side

clinical

cdisc

biostatistics

Produce clinical trial tables in R two ways: a demography Table 1 with gtsummary and an adverse-event summary with rtables — the pharmaverse package built for shell-faithful CSR layouts.

Biostatistics datasets built into R (that you can explore in the browser)

r-tutorial

biostatistics

datasets

teaching

A guided tour of the clinical and biostatistics datasets that ship with R and its packages — infert, esoph, lung, pbc, birthwt, trial and the medicaldata collection — with what each is good for, and a browser playground to try them with no install.

gtsummary for beginners: build a Table 1 from clinical data, step by step

clinical

biostatistics

gtsummary

r-tutorial

A gentle, hands-on gtsummary tutorial in R. Starting from a clinical trial dataset, we build a publication-ready Table 1 one verb at a time — then you can run it yourself in the browser.

The Markdown cheat sheet (with a live editor)

markdown

reference

A compact Markdown syntax reference: headings, emphasis, lists, links, images, code, blockquotes and tables — with the rendered result beside each rule.

Your Table 1, in one line, with gtsummary

clinical

biostatistics

gtsummary

reporting

Baseline characteristics tables are the most-copied, most-error-prone table in clinical papers. gtsummary generates a publication-ready one — with the right summary and test per variable — from a single call.

The Quarto YAML header, explained field by field

quarto

workflow

A practical tour of the Quarto YAML header: title, author, date, format, toc, theme, code-fold, execute and params — what each field does and how to avoid indentation errors.

What to know before hiring an R consultant

consulting

reproducibility

workflow

The questions that separate a one-off script from work your team can actually own: scope, reproducibility, code ownership and handover.

One source, three deliverables: Quarto to Word, PDF and HTML

quarto

reporting

How to render a single Quarto document to HTML, PDF and Word from one source file — including what each format needs and when to reach for it.

Cohen’s d and effect size in R

statistics

effect-size

r-tutorial

Why a p-value isn’t enough, and how to compute and interpret Cohen’s d effect size in R — the standardized measure of how big a difference between two groups really is.

Sensitivity, specificity, PPV and NPV in R — and why prevalence changes everything

clinical

biostatistics

diagnostics

Diagnostic accuracy from a 2×2 table in base R: sensitivity, specificity, predictive values and likelihood ratios — plus the reason the same test looks brilliant in one clinic and useless in another.

Build a website with Quarto

quarto

websites

How to turn a folder of Quarto documents into a full website — project structure, the _quarto.yml config, navigation, and one-command publishing to the web.

Linear regression in R: how to read the output

statistics

regression

r-tutorial

Fit a linear regression in R with lm() and interpret every part of the summary — coefficients, p-values, R-squared and residuals — plus a tidy table with broom.

PDF from Quarto with Typst — no LaTeX required

quarto

pdf

How to produce PDFs from Quarto using the Typst engine instead of LaTeX: faster renders, simpler setup, and clean control over page layout.

R vs Python for statistical consulting: an honest take

consulting

R

workflow

Both are excellent. For inference-heavy, publication-bound statistical work, R still has an edge — and here is the specific, non-tribal reason why.

Quarto vs R Markdown: what changed, and should you switch?

quarto

workflow

A practical comparison of Quarto and R Markdown: what’s the same, what’s genuinely new, whether your .Rmd files still work, and when it’s worth moving.

Logistic regression in R: odds ratios and ROC/AUC, done right

clinical

biostatistics

regression

pROC

A binary clinical outcome needs two things: adjusted odds ratios that answer ‘what matters?’ and an ROC/AUC that answers ‘how well does the model discriminate?’. Here is the clean R workflow for both.

Quarto code chunk options, explained

quarto

reproducibility

A practical guide to Quarto chunk options: echo, eval, include, warning, message, output, and the figure options fig-cap, fig-width and label — with a live example.

Cross-references in Quarto: figures, tables and sections

quarto

reporting

How to number and cross-reference figures, tables, equations and sections in Quarto with @fig-, @tbl- and @sec- labels, so your report’s numbering is always correct.

Chi-square test of independence in R

statistics

categorical-data

r-tutorial

Test whether two categorical variables are associated using the chi-square test in R — building the contingency table, running chisq.test, and knowing when to switch to Fisher’s exact test.

Why Welch’s t-test should be your default

statistics

simulation

A short simulation showing what unequal variances do to the classic Student’s t-test — and why the Welch correction costs you almost nothing.

Cox proportional hazards in R — and the assumption you must check

clinical

biostatistics

survival

A hazard ratio is only meaningful if the proportional-hazards assumption holds. Fit a Cox model, report hazard ratios, then test the assumption with cox.zph() — the step too many analyses skip.

One-way ANOVA in R with post-hoc tests

statistics

anova

r-tutorial

Run a one-way ANOVA in R, then use Tukey’s HSD to find which groups actually differ — with the multiple-comparison correction that keeps your error rate honest.

ROC curve and AUC in R

clinical

biostatistics

r-tutorial

How to draw a ROC curve and compute the AUC in R with the pROC package — measuring how well a model or test discriminates between two outcomes, with a confidence interval.

Power analysis before data collection — in base R

statistics

study-design

You don’t need special software to plan a sample size. power.t.test() and a power curve answer the question in ten lines of R.

Checking normality in R: Shapiro–Wilk and Q–Q plots

statistics

assumptions

r-tutorial

How to check the normality assumption in R the right way — combining the Shapiro–Wilk test with a Q–Q plot, and why the plot usually matters more than the p-value.

Analysing energy consumption time series in R

energy

time-series

forecasting

Metering data is a time series with structure hiding in plain sight. In R you can decompose electricity load into trend, weekly seasonality and noise — and read a weekday load profile straight off the data.

t-test vs ANOVA: when to use which (with R)

statistics

hypothesis-testing

anova

A t-test compares two groups; ANOVA compares three or more. See in R why they’re the same test for two groups, and why you shouldn’t run many t-tests instead of one ANOVA.

Cronbach’s alpha in R (reliability analysis)

statistics

psychometrics

r-tutorial

How to compute and interpret Cronbach’s alpha in R with the psych package — measuring the internal consistency reliability of a questionnaire scale, and what the item statistics tell you.

One template, fifty reports: parameterised Quarto in practice

quarto

reporting

workflow

How we turn a single Quarto document into a batch of per-site or per-client reports with one line of R.

Repeated-measures ANOVA in R

statistics

anova

r-tutorial

How to run a repeated-measures ANOVA in R with aov() and an Error() term — the right way to compare three or more measurements on the same subjects — with a trajectory plot.

Pearson vs Spearman correlation: which one, and when (in R)

statistics

correlation

r-tutorial

Pearson measures linear association; Spearman measures monotonic association on ranks. A clear R example shows exactly when they disagree — and which to trust.

Chi-square goodness-of-fit test in R

statistics

categorical-data

r-tutorial

Use the chi-square goodness-of-fit test in R to check whether observed counts match an expected distribution — is a die fair, do categories match expected proportions — with chisq.test.

Confidence intervals explained, with an R simulation

statistics

confidence-intervals

estimation

What a 95% confidence interval really means — demonstrated by simulating repeated samples in R — plus how to compute intervals for a mean and a proportion.

How to interpret a p-value (correctly), with R

statistics

hypothesis-testing

p-values

What a p-value is, what it isn’t, and a simulation in R that shows why p-values are uniform under the null hypothesis — the intuition most explanations skip.

Write your ggplot2 theme once, use it everywhere

ggplot2

workflow

A ten-line theme function is the cheapest branding investment an analytics team can make. Here is ours, and how to set it as a session default.

Sample size calculation in R: t-tests, proportions and correlations

power-analysis

study-design

sample-size

How to compute the sample size you need in R with power.t.test and power.prop.test — the four quantities that trade off, worked examples, and a power curve you can adapt.

Two-way ANOVA in R

statistics

anova

r-tutorial

Run and interpret a two-way ANOVA in R with aov() — two factors plus their interaction — using the built-in ToothGrowth data, with an interaction plot.

One-sample t-test in R

statistics

hypothesis-testing

r-tutorial

How to run and interpret a one-sample t-test in R with t.test() — testing whether a sample mean differs from a known value, plus the assumptions to check.

Publication-quality Kaplan–Meier curves with survival + ggplot2

survival-analysis

ggplot2

biostatistics

The survfit object has everything you need; ggplot2 does the rest. A dependency-light recipe for survival curves you control completely.

Standard deviation vs standard error: the difference, in R

statistics

r-tutorial

Standard deviation describes the spread of data; standard error describes the precision of the mean. An R simulation shows why the standard error shrinks as your sample grows.

How to read a boxplot (and make one in R)

statistics

dataviz

r-tutorial

What the box, whiskers, line and dots in a boxplot actually mean — median, quartiles, IQR and outliers — with a labelled ggplot2 example in R.

Parametric vs non-parametric tests: how to choose (in R)

statistics

r-tutorial

When to use a t-test vs a Mann–Whitney, ANOVA vs Kruskal–Wallis, Pearson vs Spearman. A clear R example shows when non-parametric tests earn their keep.

One-tailed vs two-tailed tests: which to use (with R)

statistics

hypothesis-testing

When to use a one-tailed versus a two-tailed test, why the choice must be made before seeing the data, and how each changes the p-value — shown in R.

Type I vs Type II error, explained with an R simulation

statistics

hypothesis-testing

False positives (Type I) and false negatives (Type II) are the two ways a hypothesis test can be wrong. A short R simulation makes the trade-off — and the role of power — concrete.

Z-scores explained (with R): standardising, percentiles and probabilities

statistics

distributions

r-tutorial

What a z-score is, how it maps to a percentile, and how to compute z-scores and normal probabilities in R with pnorm and qnorm — the intuition and the code.

Same correlation, wildly different data: always plot first

statistics

ggplot2

data-quality

Anscombe’s quartet in tidy form — four datasets with identical summary statistics and completely different stories. A 50-year-old lesson that still gets ignored.