Sensitivity analysis of job-training effects on reemployment for Korean women

Myoung-jae Lee, Sang Jun Lee

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

The main difficulty in treatment effect analysis with matching is accounting for unobserved differences (i.e., selection problem) between the treatment and control groups, because matching assumes no such differences. The traditional way to tackle the difficulty has been "control function" approaches with selection correction terms. This paper examines relatively new approaches: sensitivity analyses-sensitivity to unobservables-in Rosenbaum (Biometrika 74:13-26, 1987), Gastwirth et al. (Biometrika 85:907-920, 1998) and Lee (J Appl Econ 19:323-337, 2004). These sensitivity analyses are applied to the data used in Lee and Lee (J Appl Econ 20:549-562, 2005) to see how the assumption of no unobserved difference in matching affects the findings in Lee and Lee, to compare how the different sensitivity analyses perform, and to relate the "sensitivity parameters" in the different sensitivity analyses to one another. We find (i) the conclusions in Lee and Lee are weakened in the sense that only the "strong" ones survive, (ii) the sensitivity analysis in Rosenbaum (Biometrika 74:13-26, 1987) is too conservative (and inferior to Gastwirth et al.'s), and (iii) Gastwirth et al.'s and Lee's approaches agree on some findings to be insensitive, but the two approaches also disagree on some other findings.

Original languageEnglish
Pages (from-to)81-107
Number of pages27
JournalEmpirical Economics
Volume36
Issue number1
DOIs
Publication statusPublished - 2009 Jan 1

Fingerprint

Sensitivity Analysis
Parameter Sensitivity
Control Function
Treatment Effects
Training
Re-employment
Sensitivity analysis
Job training
Group
Term

Keywords

  • Job training
  • Matching
  • Sample selection
  • Sensitivity analysis

ASJC Scopus subject areas

  • Economics and Econometrics
  • Social Sciences (miscellaneous)
  • Mathematics (miscellaneous)
  • Statistics and Probability

Cite this

Sensitivity analysis of job-training effects on reemployment for Korean women. / Lee, Myoung-jae; Lee, Sang Jun.

In: Empirical Economics, Vol. 36, No. 1, 01.01.2009, p. 81-107.

Research output: Contribution to journalArticle

@article{93810f350ac64523871b7d06e0d54f60,
title = "Sensitivity analysis of job-training effects on reemployment for Korean women",
abstract = "The main difficulty in treatment effect analysis with matching is accounting for unobserved differences (i.e., selection problem) between the treatment and control groups, because matching assumes no such differences. The traditional way to tackle the difficulty has been {"}control function{"} approaches with selection correction terms. This paper examines relatively new approaches: sensitivity analyses-sensitivity to unobservables-in Rosenbaum (Biometrika 74:13-26, 1987), Gastwirth et al. (Biometrika 85:907-920, 1998) and Lee (J Appl Econ 19:323-337, 2004). These sensitivity analyses are applied to the data used in Lee and Lee (J Appl Econ 20:549-562, 2005) to see how the assumption of no unobserved difference in matching affects the findings in Lee and Lee, to compare how the different sensitivity analyses perform, and to relate the {"}sensitivity parameters{"} in the different sensitivity analyses to one another. We find (i) the conclusions in Lee and Lee are weakened in the sense that only the {"}strong{"} ones survive, (ii) the sensitivity analysis in Rosenbaum (Biometrika 74:13-26, 1987) is too conservative (and inferior to Gastwirth et al.'s), and (iii) Gastwirth et al.'s and Lee's approaches agree on some findings to be insensitive, but the two approaches also disagree on some other findings.",
keywords = "Job training, Matching, Sample selection, Sensitivity analysis",
author = "Myoung-jae Lee and Lee, {Sang Jun}",
year = "2009",
month = "1",
day = "1",
doi = "10.1007/s00181-008-0188-z",
language = "English",
volume = "36",
pages = "81--107",
journal = "Empirical Economics",
issn = "0377-7332",
publisher = "Physica-Verlag",
number = "1",

}

TY - JOUR

T1 - Sensitivity analysis of job-training effects on reemployment for Korean women

AU - Lee, Myoung-jae

AU - Lee, Sang Jun

PY - 2009/1/1

Y1 - 2009/1/1

N2 - The main difficulty in treatment effect analysis with matching is accounting for unobserved differences (i.e., selection problem) between the treatment and control groups, because matching assumes no such differences. The traditional way to tackle the difficulty has been "control function" approaches with selection correction terms. This paper examines relatively new approaches: sensitivity analyses-sensitivity to unobservables-in Rosenbaum (Biometrika 74:13-26, 1987), Gastwirth et al. (Biometrika 85:907-920, 1998) and Lee (J Appl Econ 19:323-337, 2004). These sensitivity analyses are applied to the data used in Lee and Lee (J Appl Econ 20:549-562, 2005) to see how the assumption of no unobserved difference in matching affects the findings in Lee and Lee, to compare how the different sensitivity analyses perform, and to relate the "sensitivity parameters" in the different sensitivity analyses to one another. We find (i) the conclusions in Lee and Lee are weakened in the sense that only the "strong" ones survive, (ii) the sensitivity analysis in Rosenbaum (Biometrika 74:13-26, 1987) is too conservative (and inferior to Gastwirth et al.'s), and (iii) Gastwirth et al.'s and Lee's approaches agree on some findings to be insensitive, but the two approaches also disagree on some other findings.

AB - The main difficulty in treatment effect analysis with matching is accounting for unobserved differences (i.e., selection problem) between the treatment and control groups, because matching assumes no such differences. The traditional way to tackle the difficulty has been "control function" approaches with selection correction terms. This paper examines relatively new approaches: sensitivity analyses-sensitivity to unobservables-in Rosenbaum (Biometrika 74:13-26, 1987), Gastwirth et al. (Biometrika 85:907-920, 1998) and Lee (J Appl Econ 19:323-337, 2004). These sensitivity analyses are applied to the data used in Lee and Lee (J Appl Econ 20:549-562, 2005) to see how the assumption of no unobserved difference in matching affects the findings in Lee and Lee, to compare how the different sensitivity analyses perform, and to relate the "sensitivity parameters" in the different sensitivity analyses to one another. We find (i) the conclusions in Lee and Lee are weakened in the sense that only the "strong" ones survive, (ii) the sensitivity analysis in Rosenbaum (Biometrika 74:13-26, 1987) is too conservative (and inferior to Gastwirth et al.'s), and (iii) Gastwirth et al.'s and Lee's approaches agree on some findings to be insensitive, but the two approaches also disagree on some other findings.

KW - Job training

KW - Matching

KW - Sample selection

KW - Sensitivity analysis

UR - http://www.scopus.com/inward/record.url?scp=58949090854&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=58949090854&partnerID=8YFLogxK

U2 - 10.1007/s00181-008-0188-z

DO - 10.1007/s00181-008-0188-z

M3 - Article

AN - SCOPUS:58949090854

VL - 36

SP - 81

EP - 107

JO - Empirical Economics

JF - Empirical Economics

SN - 0377-7332

IS - 1

ER -