01 / Reading Guide
Reading Guide
Econometric identification: state what makes a coefficient causal
Compare a free design-first text with two regression references, then solve an original instrumental-variable model and change its exclusion assumption.
Study route and exercise: University Reading List editorial project.
Who this book is for
For undergraduates moving from a statistics course to evaluating causal claims in observational research. It also suits analysts who can fit models but need to explain why a particular comparison answers their question. Read it before choosing an estimator, especially when a convenient dataset tempts you to skip the design stage.
Prerequisites
Basic descriptive statistics and the idea of regression help. For implementation, learn elementary data handling in one of R, Stata or Python; advanced mathematical proofs are not the starting point.
What it covers
The first part develops questions, identification and causal diagrams. The toolbox then treats designs such as matching, fixed effects, difference-in-differences, instrumental variables and regression discontinuity. The second edition includes partial identification. The emphasis is conceptual reasoning followed by implementation, with formal proof development left to more mathematical treatments.
How to use it
Draw a causal diagram for a small question before opening the code. Identify the variation you plan to use, list the assumptions that make it informative and explain which variables should or should not be controlled. Reproduce an example in one language, change a plausible assumption and describe how the causal interpretation changes. Keep an interpretation memo beside the numerical output.
Edition and sources
This record selects the 2025 second-edition paperback, ISBN 9781032580227. The author states that the publisher permits the free web version; it can receive corrections and need not share the printed book’s pagination. Data and companion materials are linked there. Publisher or author record Official online resource
Focused reading and edition boundary
The trial below uses this site's original companion problem; open its supplied assumptions and data before answering.
Reading route: 2025 second-edition route: identification, causal diagrams/paths, treatment effects and the named instrumental-variable chapter.
Try before continuing: Draw the intended Z→D→Y path and list every other way Z could affect Y. Checkpoint: Separately defend instrument independence, treatment relevance and exclusion rather than calling an instrument valid because it predicts treatment.
Edition and access: The authorized complete web edition can receive corrections independently of the paperback ISBN 9781032580227. R/Stata/Python examples are implementation choices, not alternative identification assumptions.
