Access

You are not currently logged in.

Access your personal account or get JSTOR access through your library or other institution:

login

Log in to your personal account or through your institution.

Matching as an Econometric Evaluation Estimator: Evidence from Evaluating a Job Training Programme

James J. Heckman, Hidehiko Ichimura and Petra E. Todd
The Review of Economic Studies
Vol. 64, No. 4, Special Issue: Evaluation of Training and Other Social Programmes (Oct., 1997), pp. 605-654
Published by: Oxford University Press
Stable URL: http://www.jstor.org/stable/2971733
Page Count: 50
  • Download ($42.00)
  • Subscribe ($19.50)
  • Cite this Item
Preview not available

Abstract

This paper considers whether it is possible to devise a nonexperimental procedure for evaluating a prototypical job training programme. Using rich nonexperimental data, we examine the performance of a two-stage evaluation methodology that (a) estimates the probability that a person participates in a programme and (b) uses the estimated probability in extensions of the classical method of matching. We decompose the conventional measure of programme evaluation bias into several components and find that bias due to selection on unobservables, commonly called selection bias in econometrics, is empirically less important than other components, although it is still a sizeable fraction of the estimated programme impact. Matching methods applied to comparison groups located in the same labour markets as participants and administered the same questionnaire eliminate much of the bias as conventionally measured, but the remaining bias is a considerable fraction of experimentally-determined programme impact estimates. We test and reject the identifying assumptions that justify the classical method of matching. We present a nonparametric conditional difference-in-differences extension of the method of matching that is consistent with the classical index-sufficient sample selection model and is not rejected by our tests of identifying assumptions. This estimator is effective in eliminating bias, especially when it is due to temporally-invariant omitted variables.

Page Thumbnails

  • Thumbnail: Page 
605
    605
  • Thumbnail: Page 
606
    606
  • Thumbnail: Page 
607
    607
  • Thumbnail: Page 
608
    608
  • Thumbnail: Page 
609
    609
  • Thumbnail: Page 
610
    610
  • Thumbnail: Page 
611
    611
  • Thumbnail: Page 
612
    612
  • Thumbnail: Page 
613
    613
  • Thumbnail: Page 
614
    614
  • Thumbnail: Page 
615
    615
  • Thumbnail: Page 
616
    616
  • Thumbnail: Page 
617
    617
  • Thumbnail: Page 
618
    618
  • Thumbnail: Page 
619
    619
  • Thumbnail: Page 
620
    620
  • Thumbnail: Page 
621
    621
  • Thumbnail: Page 
622
    622
  • Thumbnail: Page 
623
    623
  • Thumbnail: Page 
624
    624
  • Thumbnail: Page 
625
    625
  • Thumbnail: Page 
626
    626
  • Thumbnail: Page 
627
    627
  • Thumbnail: Page 
628
    628
  • Thumbnail: Page 
629
    629
  • Thumbnail: Page 
630
    630
  • Thumbnail: Page 
631
    631
  • Thumbnail: Page 
632
    632
  • Thumbnail: Page 
633
    633
  • Thumbnail: Page 
634
    634
  • Thumbnail: Page 
635
    635
  • Thumbnail: Page 
636
    636
  • Thumbnail: Page 
637
    637
  • Thumbnail: Page 
638
    638
  • Thumbnail: Page 
639
    639
  • Thumbnail: Page 
640
    640
  • Thumbnail: Page 
641
    641
  • Thumbnail: Page 
642
    642
  • Thumbnail: Page 
643
    643
  • Thumbnail: Page 
644
    644
  • Thumbnail: Page 
645
    645
  • Thumbnail: Page 
646
    646
  • Thumbnail: Page 
647
    647
  • Thumbnail: Page 
648
    648
  • Thumbnail: Page 
649
    649
  • Thumbnail: Page 
650
    650
  • Thumbnail: Page 
651
    651
  • Thumbnail: Page 
652
    652
  • Thumbnail: Page 
653
    653
  • Thumbnail: Page 
654
    654