Was there a riverside miracle? A hierarchical framework for evaluating programs with grouped data

Research output: Contribution to journalArticle

Abstract

This article discusses the evaluation of programs implemented at multiple sites. Two frequently used methods are pooling the data or using fixed effects (an extreme version of which estimates separate models for each site). The former approach ignores site effects. The latter incorporates site effects but lacks a framework for predicting the impact of subsequent implementations of the program (e.g., would a new implementation resemble Riverside?). I present a hierarchical model that lies between these two extremes. Using data from the Greater Avenues for Independence demonstration, I demonstrate that the model captures much of the site-to-site variation of the treatment effects but has less uncertainty than estimating the treatment effect separately for each site. I also show that when predictive uncertainty is ignored, the treatment impact for the Riverside sites is significant, but when predictive uncertainty is considered, the impact for these sites is insignificant. Finally, I demonstrate that the model extrapolates site effects with reasonable accuracy when the site being predicted does not differ substantially from the sites already observed. For example, the San Diego treatment effects could have been predicted based on their site characteristics, but the Riverside effects are consistently underpredicted.

Original languageEnglish (US)
Pages (from-to)1-11
Number of pages11
JournalJournal of Business and Economic Statistics
Volume21
Issue number1
DOIs
StatePublished - Jan 2003

Fingerprint

Grouped Data
Treatment Effects
Uncertainty
Extremes
Extrapolate
Fixed Effects
Pooling
uncertainty
Hierarchical Model
Demonstrate
Model
Framework
Grouped data
Evaluation
Estimate
Treatment effects
lack

Keywords

  • Bayesian methods
  • Predictive uncertainty
  • Site effects

ASJC Scopus subject areas

  • Statistics and Probability
  • Economics and Econometrics
  • Social Sciences (miscellaneous)
  • Statistics, Probability and Uncertainty

Cite this

@article{f545b548eb7041c4b3fdfb7999d6bdc6,
title = "Was there a riverside miracle? A hierarchical framework for evaluating programs with grouped data",
abstract = "This article discusses the evaluation of programs implemented at multiple sites. Two frequently used methods are pooling the data or using fixed effects (an extreme version of which estimates separate models for each site). The former approach ignores site effects. The latter incorporates site effects but lacks a framework for predicting the impact of subsequent implementations of the program (e.g., would a new implementation resemble Riverside?). I present a hierarchical model that lies between these two extremes. Using data from the Greater Avenues for Independence demonstration, I demonstrate that the model captures much of the site-to-site variation of the treatment effects but has less uncertainty than estimating the treatment effect separately for each site. I also show that when predictive uncertainty is ignored, the treatment impact for the Riverside sites is significant, but when predictive uncertainty is considered, the impact for these sites is insignificant. Finally, I demonstrate that the model extrapolates site effects with reasonable accuracy when the site being predicted does not differ substantially from the sites already observed. For example, the San Diego treatment effects could have been predicted based on their site characteristics, but the Riverside effects are consistently underpredicted.",
keywords = "Bayesian methods, Predictive uncertainty, Site effects",
author = "Rajeev Dehejia",
year = "2003",
month = "1",
doi = "10.1198/073500102288618702",
language = "English (US)",
volume = "21",
pages = "1--11",
journal = "Journal of Business and Economic Statistics",
issn = "0735-0015",
publisher = "American Statistical Association",
number = "1",

}

TY - JOUR

T1 - Was there a riverside miracle? A hierarchical framework for evaluating programs with grouped data

AU - Dehejia, Rajeev

PY - 2003/1

Y1 - 2003/1

N2 - This article discusses the evaluation of programs implemented at multiple sites. Two frequently used methods are pooling the data or using fixed effects (an extreme version of which estimates separate models for each site). The former approach ignores site effects. The latter incorporates site effects but lacks a framework for predicting the impact of subsequent implementations of the program (e.g., would a new implementation resemble Riverside?). I present a hierarchical model that lies between these two extremes. Using data from the Greater Avenues for Independence demonstration, I demonstrate that the model captures much of the site-to-site variation of the treatment effects but has less uncertainty than estimating the treatment effect separately for each site. I also show that when predictive uncertainty is ignored, the treatment impact for the Riverside sites is significant, but when predictive uncertainty is considered, the impact for these sites is insignificant. Finally, I demonstrate that the model extrapolates site effects with reasonable accuracy when the site being predicted does not differ substantially from the sites already observed. For example, the San Diego treatment effects could have been predicted based on their site characteristics, but the Riverside effects are consistently underpredicted.

AB - This article discusses the evaluation of programs implemented at multiple sites. Two frequently used methods are pooling the data or using fixed effects (an extreme version of which estimates separate models for each site). The former approach ignores site effects. The latter incorporates site effects but lacks a framework for predicting the impact of subsequent implementations of the program (e.g., would a new implementation resemble Riverside?). I present a hierarchical model that lies between these two extremes. Using data from the Greater Avenues for Independence demonstration, I demonstrate that the model captures much of the site-to-site variation of the treatment effects but has less uncertainty than estimating the treatment effect separately for each site. I also show that when predictive uncertainty is ignored, the treatment impact for the Riverside sites is significant, but when predictive uncertainty is considered, the impact for these sites is insignificant. Finally, I demonstrate that the model extrapolates site effects with reasonable accuracy when the site being predicted does not differ substantially from the sites already observed. For example, the San Diego treatment effects could have been predicted based on their site characteristics, but the Riverside effects are consistently underpredicted.

KW - Bayesian methods

KW - Predictive uncertainty

KW - Site effects

UR - http://www.scopus.com/inward/record.url?scp=0037237264&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037237264&partnerID=8YFLogxK

U2 - 10.1198/073500102288618702

DO - 10.1198/073500102288618702

M3 - Article

VL - 21

SP - 1

EP - 11

JO - Journal of Business and Economic Statistics

JF - Journal of Business and Economic Statistics

SN - 0735-0015

IS - 1

ER -