Matching methods for selection of participants for follow-up

Elizabeth Stuart, Nicholas S. Lalongo

Research output: Contribution to journalArticle

Abstract

This work examines ways to make the best use of limited resources when selecting individuals to follow up in a longitudinal study estimating causal effects. In the setting under consideration, covariate information is available for all individuals but outcomes have not yet been collected and may be expensive to gather, and thus only a subset of the comparison participants are followed. Expressions in Rubin and Thomas (1996) show the benefits that can be obtained, in terms of reduced bias and variance of the estimated treatment effect, of selecting comparison individuals well matched to those in the treated group compared with a random sample of comparison individuals. We primarily consider nonexperimental settings but also consider implications for randomized trials. The methods are illustrated using data from the Johns Hopkins University Baltimore Prevention Program, which included data collection from age 6 to young adulthood of participants in an evaluation of 2 early elementary-school-based universal prevention programs.

Original languageEnglish (US)
Pages (from-to)746-765
Number of pages20
JournalMultivariate Behavioral Research
Volume45
Issue number4
DOIs
StatePublished - 2010

Fingerprint

Baltimore
Longitudinal Studies
Causal Effect
Randomized Trial
Longitudinal Study
Treatment Effects
Covariates
Resources
Subset
Evaluation

ASJC Scopus subject areas

  • Experimental and Cognitive Psychology
  • Statistics and Probability

Cite this

Matching methods for selection of participants for follow-up. / Stuart, Elizabeth; Lalongo, Nicholas S.

In: Multivariate Behavioral Research, Vol. 45, No. 4, 2010, p. 746-765.

Research output: Contribution to journalArticle

@article{b57fa82538f3488582eef3a3b98dcd9d,
title = "Matching methods for selection of participants for follow-up",
abstract = "This work examines ways to make the best use of limited resources when selecting individuals to follow up in a longitudinal study estimating causal effects. In the setting under consideration, covariate information is available for all individuals but outcomes have not yet been collected and may be expensive to gather, and thus only a subset of the comparison participants are followed. Expressions in Rubin and Thomas (1996) show the benefits that can be obtained, in terms of reduced bias and variance of the estimated treatment effect, of selecting comparison individuals well matched to those in the treated group compared with a random sample of comparison individuals. We primarily consider nonexperimental settings but also consider implications for randomized trials. The methods are illustrated using data from the Johns Hopkins University Baltimore Prevention Program, which included data collection from age 6 to young adulthood of participants in an evaluation of 2 early elementary-school-based universal prevention programs.",
author = "Elizabeth Stuart and Lalongo, {Nicholas S.}",
year = "2010",
doi = "10.1080/00273171.2010.503544",
language = "English (US)",
volume = "45",
pages = "746--765",
journal = "Multivariate Behavioral Research",
issn = "0027-3171",
publisher = "Psychology Press Ltd",
number = "4",

}

TY - JOUR

T1 - Matching methods for selection of participants for follow-up

AU - Stuart, Elizabeth

AU - Lalongo, Nicholas S.

PY - 2010

Y1 - 2010

N2 - This work examines ways to make the best use of limited resources when selecting individuals to follow up in a longitudinal study estimating causal effects. In the setting under consideration, covariate information is available for all individuals but outcomes have not yet been collected and may be expensive to gather, and thus only a subset of the comparison participants are followed. Expressions in Rubin and Thomas (1996) show the benefits that can be obtained, in terms of reduced bias and variance of the estimated treatment effect, of selecting comparison individuals well matched to those in the treated group compared with a random sample of comparison individuals. We primarily consider nonexperimental settings but also consider implications for randomized trials. The methods are illustrated using data from the Johns Hopkins University Baltimore Prevention Program, which included data collection from age 6 to young adulthood of participants in an evaluation of 2 early elementary-school-based universal prevention programs.

AB - This work examines ways to make the best use of limited resources when selecting individuals to follow up in a longitudinal study estimating causal effects. In the setting under consideration, covariate information is available for all individuals but outcomes have not yet been collected and may be expensive to gather, and thus only a subset of the comparison participants are followed. Expressions in Rubin and Thomas (1996) show the benefits that can be obtained, in terms of reduced bias and variance of the estimated treatment effect, of selecting comparison individuals well matched to those in the treated group compared with a random sample of comparison individuals. We primarily consider nonexperimental settings but also consider implications for randomized trials. The methods are illustrated using data from the Johns Hopkins University Baltimore Prevention Program, which included data collection from age 6 to young adulthood of participants in an evaluation of 2 early elementary-school-based universal prevention programs.

UR - http://www.scopus.com/inward/record.url?scp=77955909187&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955909187&partnerID=8YFLogxK

U2 - 10.1080/00273171.2010.503544

DO - 10.1080/00273171.2010.503544

M3 - Article

C2 - 21221424

AN - SCOPUS:77955909187

VL - 45

SP - 746

EP - 765

JO - Multivariate Behavioral Research

JF - Multivariate Behavioral Research

SN - 0027-3171

IS - 4

ER -