Article 2012

Use of propensity score matching for training research with observational data

GS
Gemici, Sinan
 RJ
Rojewski, Jay W.
 LI
Lee, In Heok
International Journal of Training Research
Journal · Vol. 10 · Issue 3 · pp. 219-232
Abstract

Evaluations of vocational education and training (VET) programs play a key role in informing training policy in Australia and elsewhere. Increasingly, such evaluations use observational data from surveys or administrative collections to assess the effectiveness of VET programs and interventions. The difficulty associated with using observational data is that they are inherently prone to selection bias, which results from individuals self-selecting into a given VET program based on differences in background characteristics or other external factors. The effects of the VET program on outcomes of interest are thus confounded with the effects of pre-existing systematic differences between program participants and non-participants. Propensity score matching (PSM) can mitigate selection bias in evaluation studies with observational data by statistically balancing program participants and non-participants post hoc on observed background characteristics. This article seeks to offer a general introduction to PSM and to provide interested VET researchers with an initial stepping stone for using the method in their own work. © eContent Management Pty Ltd.

Keywords

Author Keywords

vocational education and training Program evaluation selection bias propensity score matching Observational data

Index Keywords

Author Affiliations
Program of Workforce Education, University of Georgia, Athens, GA, United States
National Centre for Vocational Education Research, Adelaide, SA, Australia
Funding & Acknowledgements
No funding information
References 10 References
1 Propensity Score A Means to an End, (2001)
2 Austin, Peter C., The relative ability of different propensity score methods to balance measured covariates between treated and untreated subjects in observational studies, Medical Decision Making, 29, 6, pp. 661-677, (2009)
3 Austin, Peter C., A comparison of the ability of different propensity score models to balance measured variables between treated and untreated subjects: A Monte Carlo study, Statistics in Medicine, 26, 4, pp. 734-753, (2007)
4 Becker, Sascha O., Sensitivity analysis for average treatment effects, Stata Journal, 7, 1, pp. 71-83, (2007)
5 Caliendo, Marco, Some practical guidance for the implementation of propensity score matching, Journal of Economic Surveys, 22, 1, pp. 31-72, (2008)
6 Analysis of Binary Data, (1970)
7 Cox-Edwards, Alejandra, Remittances and Labor Force Participation in Mexico: An Analysis Using Propensity Score Matching, World Development, 37, 5, pp. 1004-1014, (2009)
8 Curtis, Lesley H., Using inverse probability-weighted estimators in comparative effectiveness analyses with observational databases, Medical Care, 45, 10 SUPPL. 2, pp. S103-S107, (2007)
9 D’Agostino, Ralph B., Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group, Statistics in Medicine, 17, 19, pp. 2265-2281, (1998)
10 de Anda, Diane, Intervention research and program evaluation in the school setting: Issues and alternative research designs, Children and Schools, 29, 2, pp. 87-94, (2007)
Quick Actions
Full Text via DOI
Citation Metrics
11
Times Cited (Scopus)
References 10
Document Identifiers