Skip to contents

This hypothetical dataset shows a realized assignment (Z) and the corresponding revealed outcome (Y). Unobserved treated (Y_Z_1) and untreated (Y_Z_0) potential outcomes are marked as NA, as is the village-level treatment effect tau_i.

Usage

table_2_2

Format

A data frame with 7 rows and 6 columns:

Village

Village number, 1 to 7

Z

Realized treatment assignment, 1 for the two treated villages

Y_Z_0

Untreated potential outcome, NA where treatment was assigned

Y_Z_1

Treated potential outcome, NA where control was assigned

Y

Revealed outcome

tau_i

Village-level treatment effect, NA throughout: it is never observable

Source

Derived from Table 2.1 (file 42291) by applying the assignment printed in the book. Built by data-raw/book_tables.R.