Book contents
- Frontmatter
- Contents
- List of Figures
- List of Tables
- List of Boxes
- List of Contributors
- Acknowledgements
- 1 A New Era of Experimental Political Science
- Part I Experimental Designs
- Part II Experimental Data
- Part III Experimental Treatments and Measures
- Part IV Experimental Analys is and Presentation
- 14 Advances in Experimental Mediation Analysis
- 15 Subgroup Analysis: Pitfalls, Promise, and Honesty
- 16 Spillover Effects in Experimental Data
- 17 Visualize as You Randomize: Design-Based Statistical Graphs for Randomized Experiments
- Part V Experimental Reliability and Generalizability
- Part VI Using Experiments to study Identity
- Part VII Using Experiments to Study Government Actions
- Author Index
- Subject Index
14 - Advances in Experimental Mediation Analysis
from Part IV - Experimental Analys is and Presentation
Published online by Cambridge University Press: 08 March 2021
- Frontmatter
- Contents
- List of Figures
- List of Tables
- List of Boxes
- List of Contributors
- Acknowledgements
- 1 A New Era of Experimental Political Science
- Part I Experimental Designs
- Part II Experimental Data
- Part III Experimental Treatments and Measures
- Part IV Experimental Analys is and Presentation
- 14 Advances in Experimental Mediation Analysis
- 15 Subgroup Analysis: Pitfalls, Promise, and Honesty
- 16 Spillover Effects in Experimental Data
- 17 Visualize as You Randomize: Design-Based Statistical Graphs for Randomized Experiments
- Part V Experimental Reliability and Generalizability
- Part VI Using Experiments to study Identity
- Part VII Using Experiments to Study Government Actions
- Author Index
- Subject Index
Summary
Mediation analysis has been called \harder than it looks" (Bullock and Ha 2011) due to difficulties of experimental identification. However, recent work has clarified that, while hard, some experimental designs for mediation can be informative. Other recent work has provided “easier” substitutes for mediation analysis. This chapter has two goals. First, to summarize some of the findings published since Bullock and Ha (2011) and to consider the implications these findings have for mediation analysis. Second, to consider the situations in which a close alternative to mediation analysis would be useful (either as a supplement or a substitute). Such situations often depend on the motivation for the analysis.
Keywords
- Type
- Chapter
- Information
- Advances in Experimental Political Science , pp. 257 - 270Publisher: Cambridge University PressPrint publication year: 2021
- 8
- Cited by