Direct and indirect effects in a logit model
Open accessAbstract
In this article, I discuss a method by Erikson et al. (2005, Proceedings of the National Academy of Science 102: 9730–9733) for decomposing a total effect in a logit model into direct and indirect effects. Moreover, I extend this method in three ways. First, in the original method the variable through which the indirect effect occurs is assumed to be normally distributed. In this article, the method is generalized by allowing this variable to have any distribution. Second, the original method did not provide standard errors for the estimates. In this article, the bootstrap is proposed as a method of providing those. Third, I show how to include control variables in this decomposition, which was not allowed in the original method. The original method and these extensions are implemented in the ldecomp command.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Cites
An Introduction to the Bootstrap
Efron, B., & Tibshirani, R. J. (1994) · doi:10.1201/9780429246593 · Cited by 40,500 · see citing works
An Introduction to the Bootstrap
Efron, B., & Tibshirani, R. J. (1993) · doi:10.1007/978-1-4899-4541-9 · Cited by 21,230 · see citing works
Eulau, H. H. F. (1960) · American Political Science Review · doi:10.2307/1952653 · Cited by 4,749 · see citing works
Beyond the Running Tally: Partisan Bias in Political Perceptions
Bartels, L. M. (2002) · Political Behavior · doi:10.1023/a:1021226224601 · Cited by 1,468 · see citing works
An extension of the Blinder-Oaxaca decomposition technique to logit and probit models
Fairlie, R. W. (2005) · Journal of Economic and Social Measurement · doi:10.3233/jem-2005-0259 · Cited by 1,096 · see citing works
Decomposing differences in the first moment
Yun, M. (2003) · Economics Letters · doi:10.1016/j.econlet.2003.09.008 · Cited by 517 · see citing works
Estimation of Marginal Effects using Margeff
Bartus, T. (2005) · The Stata Journal Promoting communications on statistics and Stata · doi:10.1177/1536867x0500500303 · Cited by 465 · see citing works
On class differentials in educational attainment
Erikson, R. S., Goldthorpe, J. H., Jackson, M., Yaish, M., & Cox, D. R. (2005) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.0502433102 · Cited by 352 · see citing works
Primary and Secondary Effects in Class Differentials in Educational Attainment
Jackson, M., Erikson, R., Goldthorpe, J., & Yaish, M. (2007) · Acta Sociologica · doi:10.1177/0001699307080926 · Cited by 293 · see citing works
An extension of the Blinder–Oaxaca decomposition to nonlinear models
Bauer, T., & Sinning, M. (2008) · AStA Advances in Statistical Analysis · doi:10.1007/s10182-008-0056-3 · Cited by 264 · see citing works
The Blinder–Oaxaca Decomposition for Nonlinear Regression Models
Sinning, M., Hahn, M. H., & Bauer, T. (2008) · The Stata Journal Promoting communications on statistics and Stata · doi:10.1177/1536867x0800800402 · Cited by 245 · see citing works
The Employment of Married Women in the United Kingdom 1970-83
Gomulka, J., & Stern, N. H. (1990) · Economica · doi:10.2307/2554159 · Cited by 154 · see citing works
Plant size and the decline of unionism
Even, W. E., & Macpherson, D. A. (1990) · Economics Letters · doi:10.1016/0165-1765(90)90035-y · Cited by 133 · see citing works
FAIRLIE: Stata module to generate nonlinear decomposition of binary outcome differentialsPreprint
Jann, B. (2006) · RePEc: Research Papers in Economics · doi:10.7892/boris.69420 · Cited by 102 · see citing works
Predict and Adjust with Logistic Regression
Buis, M. L. (2007) · The Stata Journal Promoting communications on statistics and Stata · doi:10.1177/1536867x0700700206 · Cited by 22 · see citing works
Self-citationStata Tip 48: Discrete Uses for Uniform()
Buis, M. L. (2007) · The Stata Journal Promoting communications on statistics and Stata · doi:10.1177/1536867x0700700309 · Cited by 9 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
Myers, B., Lombard, C., Joska, J. A., Abdullah, F., Naledi, T., Lund, C., Petersen Williams, P., Stein, D. J., & Sorsdahl, K. R. (2021) · AIDS and Behavior · doi:10.1007/s10461-021-03263-3 · Cited by 47 · see citing works
Skafida, V., & Devaney, J. (2023) · Child Abuse & Neglect · doi:10.1016/j.chiabu.2022.105991 · Cited by 16 · see citing works
Li, Y., Graubard, B. I., Huang, P., & Gastwirth, J. L. (2015) · Statistics in Medicine · doi:10.1002/sim.6357 · Cited by 10 · see citing works
- Authors
- Buis, M. L.
- Affiliations
- University of Tübingen (DE)
- Corresponding author
- Buis, Maarten L.
- ORCID
- Buis, Maarten L.
- Venue
- The Stata Journal: Promoting communications on statistics and Stata
- Publisher
- SAGE Publications
- Published
- 2010-03
- Volume
- 10
- Issue
- 1
- Pages
- 11-29
- ISSN
- 1536-867X (ISSN-L)
- 1536-8734
- PMC
- PMC3314333
- OpenAlex
- W1599214905
- Cited by
- 251
- References
- 16
- Citations by year
- 2026: 12 · 2025: 15 · 2024: 19 · 2023: 30 · 2022: 24 · 2021: 22 · 2020: 26 · 2019: 16 · 2018: 15 · 2017: 13 · 2016: 12 · 2015: 5 · 2014: 11 · 2013: 9 · 2012: 8
- FWCI
- 8.46
- Type
- journal article
- Language
- en
- Open access
- bronze
- doi.org
- Locations
- The Stata Journal Promoting communications on statistics and Stata (publishedVersion)
- PubMed (publishedVersion)
- PubMed Central (acceptedVersion)
- Funders
- National Institutes of Health (5p01ag021079-08)
- National Institute on Aging (P01-AG021079)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Spatial and Panel Data Analysis
- Keywords
- direct and indirect effects, logit model, effect decomposition, mediation analysis, bootstrap standard errors, control variables
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
