This namespace is for what these methods do to web-measurement data — a sample that is a ranking list, a site that shares a tag manager with two hundred others, a crawl that generates hypotheses the way it generates rows, a hand-coded ground truth, a plan deposited before the data. It is not a statistics textbook. A general account of a t-test, a logit, or Cohen's kappa belongs in one; what belongs here is the unit-of-analysis error, the family that a crawl invents, and the reporting gaps this corpus actually has. The publication corpus behind these pages is seven venues (CCS, IMC, NDSS, PETS, USENIX Security, TheWebConf, IEEE S&P, 2010–2026, 5,859 extracted papers). 1,762 ran statistical inference; 1,357 recruited human participants; 3,318 hand-coded something. Each child names its own population.

Annotation and inter-rater agreement also compose, and the split between them is deliberate: agreement owns the coefficients and the reliability of hand coding, annotation owns the validation design and the obligations an LLM annotator inherits from being a classifier. Hypothesis testing and p-value corrections share a population on purpose (papers that ran a test), so they compose. Regression is the “by how much, holding other things constant” question; the child is about the forms that takes on web-measurement rows. Biases is the page that names what the Design choices did to the number. Inter-rater agreement is for the slice that stopped being automatic. Preregistration is the plan, not the artifact (Artifacts) and not the ethics review (Ethics). How many sites is the one page here that runs before any of the others: it is about choosing n, which Sampling assumes you have already done and Hypothesis testing assumes you cannot change.

• Sampling / Website selection / Crawling location / Longitudinal — choosing the frame, the vantage, the pin. Biases measures what those choices cost; it does not choose them.
• User studies — recruiting people. Regression and preregistration are often user-study methods that crawl papers borrow.
• Literature review — the keyword-derived denominator. A related-work count is not a statistical estimator, but it has the same missing-denominator failure mode.
• Public relations — the sentence that travels. The denominator has to be in it.

The 5,859 / 1,762 / 1,357 / 3,318 are paper counts from the 5,859-paper extraction (seven venues, 2010–2026). 2025–2026 venue-years are provisional — see corpus. The 8 is a wiki-page count as of 2026-09-11. Queries: statistics. Joint sitting: design.