NO. 37 · Computational & Data

Statistics for Researchers

The statistics your reviewers assume you know. Probability, estimation, honest testing, calibrated intervals, regression, and causal inference, built for people whose results have to survive peer review.

You can choose and defend an estimator, state exactly what your p-value and interval do and do not claim, design a study with adequate power, and reason about causation with DAGs instead of hope.

26 competencies · 6 interactive widget challenges · 24 to 34 hours of guided study
For researchers in any field who touch data

Probability that pays rent

Axioms and random variables

Every statistical claim you will ever make is a sentence in this language; learn the grammar before the rhetoric.

Joint distributions and expectations

Conditional and marginal thinking is where most published confusion starts, and expectation is the accountant of the whole subject.

The catalog and the law of large numbers

Knowing which distribution you are holding, and why averages settle down, is the difference between modeling and matching shapes.

The CLT and simulation

The central limit theorem is why any of this works on real data, and simulation is how you check your intuition without waiting for a proof.

Estimation

Estimators and the method of moments

Bias, variance, and consistency are the report card every estimator carries; method of moments is the first honest attempt.

Maximum likelihood and Fisher information

MLE is the workhorse of applied statistics, and the Cramer-Rao bound tells you when to stop looking for something better.

Bias-variance and sampling distributions

The standard error is the most-quoted, least-understood number in science; here is where it actually comes from.

Bootstrap and robust estimationwidget challenge

Resampling gives you uncertainty when formulas run out, and M-estimators keep one outlier from owning your conclusion.

Testing, honestly

Neyman-Pearson and powerwidget challenge

Type I, Type II, and power are the terms of the bet every test places; most underpowered studies never knew they were betting.

Classical tests and the p-valuewidget challenge

The t, chi-square, and F tests done by hand once each, and what a p-value is and is not, which is the most misstated fact in research.

Multiple testing and forking paths

Twenty tests buy one false discovery at full price; FWER, FDR, and preregistration are how you stop paying it.

Equivalence and the replication crisis

Absence of evidence has its own test, and the replication crisis is the case study in what happens when a field skips this stage.

Intervals and calibration

Confidence intervals, exact and bootstrap

An interval is a procedure with a coverage promise; exact, asymptotic, and bootstrap constructions keep that promise differently.

Profile likelihood and prediction intervals

Confidence intervals speak about parameters, prediction intervals about the next observation; confusing them flatters every forecast.

Calibration and honest communicationwidget challenge

When 95 percent really means 95 percent, and how to draw uncertainty so the reader hears what you actually know.

Regression

OLS as geometry, and its assumptions

Least squares is a projection, and each assumption is a load bearing wall; know which one fails before the estimate leans.

Diagnostics and heteroscedasticity

Residuals, leverage, and influence are the regression's own testimony about itself; learn to take the deposition.

Robust fits and interactions

Real data have outliers and real effects have modifiers; robust regression and interaction terms handle both without theater.

Model selection and the causal warning

AIC, BIC, and cross-validation pick predictive models; none of them make a coefficient causal, and the warning label is part of the method.

Causal inference

Potential outcomes and RCTs

The fundamental problem of causal inference has exactly one clean solution, and it is randomization done right.

Confounding and DAGswidget challenge

A directed graph turns arguments about bias into arithmetic; the backdoor criterion is worth a hundred seminar fights.

Instruments and discontinuities

When you cannot randomize, instruments and cutoffs are the two most credible substitutes, each with a price printed in assumptions.

Difference-in-differences and sensitivity

Parallel trends is an assumption you can almost see, and sensitivity analysis bounds what the confounder you missed could have done.

Capstones

Capstone: a designed RCT, end to endwidget challenge

Design, power, preregister, run, analyze, report: the whole discipline in one study you could actually submit.

Capstone: an observational study

Confounding is the default state of nature; this capstone is the drill for concluding carefully anyway.

The research-analyst workflow card

The whole path compressed onto one card you can pin above the desk where the analysis actually happens.

This page is prerendered for SEO and accessibility. With JavaScript, it hydrates into the live guided path: placement quiz, spaced practice, and interactive widget challenges.