What is included with this book?
Preface | p. xv |
Introduction | p. 1 |
Drawing Inferences from Intuitions, Anecdotes, and Correlations | p. 2 |
Experiments as a Solution to the Problem of Unobserved Confounders | p. 5 |
Experiments as Fair Tests | p. 7 |
Field Experiments | p. 8 |
Advantages and Disadvantages of Experimenting in Real-World Settings | p. 13 |
Naturally Occurring Experiments and Quasi-Experiments | p. 15 |
Plan of the Book | p. 17 |
Suggested Readings | p. 18 |
Exercises | p. 18 |
Causal Inference and Experimentation | p. 21 |
Potential Outcomes | p. 21 |
Average Treatment Effects | p. 23 |
Random Sampling and Expectations | p. 26 |
Random Assignment and Unbiased Inference | p. 30 |
The Mechanics of Random Assignment | p. 36 |
The Threat of Selection Bias When Random Assignment Is Not Used | p. 37 |
Two Core Assumptions about Potential Outcomes | p. 39 |
Excludability | p. 39 |
Non-interference | p. 43 |
Summary | p. 44 |
Suggested Readings | p. 46 |
Exercises | p. 46 |
Sampling Distributions, Statistical Inference, and Hypothesis Testing | p. 51 |
Sampling Distributions | p. 52 |
The Standard Error as a Measure of Uncertainty | p. 54 |
Estimating Sampling Variability 59 3.4- Hypothesis Testing | p. 61 |
Confidence Intervals | p. 66 |
Sampling Distributions for Experiments That Use Block or Cluster Random Assignment | p. 71 |
Block Random Assignment | p. 71 |
Matched Pair Design | p. 77 |
Summary of the Advantages and Disadvantages of Blocking | p. 79 |
Cluster Random Assignment | p. 80 |
Summary | p. 85 |
Suggested Readings | p. 86 |
Exercises | p. 86 |
Power | p. 93 |
Using Covariates in Experimental Design and Analysis | p. 95 |
Using Covariates to Rescale Outcomes | p. 96 |
Adjusting for Covariates Using Regression | p. 102 |
Covariate Imbalance and the Detection of Administrative Errors | p. 105 |
Blocked Randomization and Covariate Adjustment | p. 109 |
Analysis of Block Randomized Experiments with Treatment Probabilities That Vary by Block | p. 116 |
Summary | p. 121 |
Suggested Readings | p. 123 |
Exercises | p. 123 |
One-Sided Noncompliance | p. 131 |
New Definitions and Assumptions | p. 134 |
Denning Causal Effects for the Case of One-Sided Noncompliance 13 | p. 137 |
The Non-interference Assumption for Experiments That Encounter Noncompliance | p. 138 |
The Excludability Assumption for One-Sided Noncompliance | p. 140 |
Average Treatment Effects, Intent-to-Treat Effects, and Complier Average Causal Effects | p. 141 |
Identification of the CACE | p. 143 |
Estimation | p. 149 |
Avoiding Common Mistakes | p. 152 |
Evaluating the Assumptions Required to Identify the CACE 15 | p. 155 |
Non-interference Assumption | p. 155 |
Exclusion Restriction | p. 156 |
Statistical Inference | p. 157 |
Designing Experiments in Anticipation of Noncompliance | p. 161 |
Estimating Treatment Effects When Some Subjects Receive "Partial Treatment" | p. 164 |
Summary | p. 165 |
Suggested Readings | p. 167 |
Exercises | p. 168 |
Two-Sided Noncompliance | p. 173 |
Two-Sided Noncompliance: New Definitions and Assumptions | p. 175 |
ITT, ITT_{D}, and CACE under Two-Sided Noncompliance | p. 179 |
A Numerical Illustration of the Role of Monotonicity | p. 181 |
Estimation of the CACE: An Example | p. 185 |
Discussion of Assumptions | p. 189 |
Monotonicity | p. 190 |
Exclusion Restriction | p. 191 |
Random Assignment | p. 192 |
Design Suggestions | p. 192 |
Downstream Experimentation | p. 193 |
Summary | p. 204 |
Suggested Readings | p. 206 |
Exercises | p. 206 |
Attrition | p. 211 |
Conditions Under Which Attrition Leads to Bias | p. 215 |
Special Forms of Attrition | p. 219 |
Redefining the Estimand When Attrition Is Not a Function of Treatment Assignment | p. 224 |
Placing Bounds on the Average Treatment Effect | p. 226 |
Addressing Attrition: An Empirical Example | p. 230 |
Addressing Attrition with Additional Data Collection | p. 236 |
Two Frequently Asked Questions | p. 241 |
Summary | p. 243 |
Suggested Readings | p. 244 |
Exercises | p. 244 |
Optimal Sample Allocation for Second-Round Sampling | p. 248 |
Interference between Experimental Units | p. 253 |
Identifying Causal Effects in the Presence of Localized Spillover | p. 256 |
Spatial Spillover | p. 260 |
Using Nonexperimental Units to Investigate Spillovers | p. 264 |
An Example of Spatial Spillovers in Two Dimensions | p. 264 |
Within-Subjects Design and Time-Series Experiments | p. 273 |
Waitlist Designs (Also Known as Stepped-Wedge Designs) | p. 276 |
Summary | p. 281 |
Suggested Readings | p. 283 |
Exercises | p. 283 |
Heterogeneous Treatment Effects | p. 289 |
Limits to What Experimental Data Tell Us about Treatment Effect Heterogeneity | p. 291 |
Bounding Var (¿) and Testing for Heterogeneity | p. 292 |
Two Approaches to the Exploration of Heterogeneity: Covariates and Design | p. 296 |
Assessmg Treatment-by-Covariate Interactions | p. 296 |
Caution Is Required When Interpreting Treatment-by-Covariate Interactions | p. 299 |
Assessing Treatment-by-Treatment Interactions | p. 303 |
Using Regression to Model Treatment Effect Heterogeneity | p. 305 |
Automating the Search for Interactions | p. 310 |
Summary | p. 310 |
Suggested Readings | p. 312 |
Exercises | p. 313 |
Mediation | p. 319 |
Regression-Based Approaches to Mediation | p. 322 |
Mediation Analysis from a Potential Outcomes Perspective | p. 325 |
Why Experimental Analysis of Mediators Is Challenging | p. 328 |
Ruling Out Mediators? | p. 330 |
What about Experiments That Manipulate the Mediator? | p. 331 |
Implicit Mediation Analysis | p. 333 |
Summary | p. 336 |
Suggested Readings | p. 338 |
Exercises | p. 338 |
Treatment Postcards Mailed to Michigan Households | p. 343 |
Integration of Research Findings | p. 347 |
Estimation of Population Average Treatment Effects | p. 350 |
A Bayesian Framework for Interpreting Research Findings | p. 353 |
Replication and Integration of Experimental Findings: An Example | p. 358 |
Treatments That Vary in Intensity: Extrapolation and Statistical Modeling | p. 366 |
Summary | p. 377 |
Suggested Readings | p. 378 |
Exercises | p. 379 |
Instructive Examples of Experimental Design | p. 383 |
Using Experimental Design to Distinguish between Competing Theories | p. 384 |
Oversampling Subjects Based on Their Anticipated Response to Treatment | p. 387 |
Comprehensive Measurement of Outcomes | p. 393 |
Factorial Design and Special Cases of Non-interference | p. 395 |
Design and Analysis of Experiments In Which Treatments Vary with Subjects' Characteristics | p. 400 |
Design and Analysis of Experiments In Which Failure to Receive Treatment Has a Causal Effect | p. 406 |
Addressing Complications Posed by Missing Data | p. 410 |
Summary | p. 414 |
Suggested Readings | p. 415 |
Exercises | p. 416 |
Writing a Proposal, Research Report, and Journal Article | p. 425 |
Writing the Proposal | p. 426 |
Writing the Research Report | p. 435 |
Writing the Journal Article | p. 440 |
Archiving Data | p. 442 |
Summary | p. 444 |
Suggested Readings | p. 445 |
Exercises | p. 445 |
Protection of Human Subjects | p. 447 |
Regulatory Guidelines | p. 447 |
Guidelines for Keeping Field Experiments within Regulatory Boundaries | p. 449 |
Suggested Field Experiments for Class Projects | p. 453 |
Crafting Your Own Experiment | p. 453 |
Suggested Experimental Topics for Practicum Exercises | p. 455 |
References | p. 461 |
Index | p. 479 |
Table of Contents provided by Ingram. All Rights Reserved. |