Hostname: page-component-669899f699-tpknm Total loading time: 0 Render date: 2025-04-26T11:11:46.674Z Has data issue: false hasContentIssue false
  • English
  • Français

A Class of Distribution-Free Models for Longitudinal Mediation Analysis

Published online by Cambridge University Press:  01 January 2025

D. Gunzler ,
W. Tang ,
N. Lu ,
P. Wu  and
X. M. Tu
D. Gunzler*
Affiliation:
Case Western Reserve University at MetroHealth Medical Center
W. Tang
Affiliation:
University of Rochester
N. Lu
Affiliation:
University of Rochester and Canandaigua VA Medical Center
P. Wu
Affiliation:
Christiana Care Health System
X. M. Tu
Affiliation:
University of Rochester and Canandaigua VA Medical Center
*
Requests for reprints should be sent to D. Gunzler, Center for Health Care Research & Policy, Case Western Reserve University at MetroHealth Medical Center, 2500 MetroHealth Drive, Cleveland, OH 44109-1998, USA. E-mail: dgunzler@metrohealth.org
Get access Rights & Permissions [Opens in a new window]

Abstract

Mediation analysis constitutes an important part of treatment study to identify the mechanisms by which an intervention achieves its effect. Structural equation model (SEM) is a popular framework for modeling such causal relationship. However, current methods impose various restrictions on the study designs and data distributions, limiting the utility of the information they provide in real study applications. In particular, in longitudinal studies missing data is commonly addressed under the assumption of missing at random (MAR), where current methods are unable to handle such missing data if parametric assumptions are violated.

In this paper, we propose a new, robust approach to address the limitations of current SEM within the context of longitudinal mediation analysis by utilizing a class of functional response models (FRM). Being distribution-free, the FRM-based approach does not impose any parametric assumption on data distributions. In addition, by extending the inverse probability weighted (IPW) estimates to the current context, the FRM-based SEM provides valid inference for longitudinal mediation analysis under the two most popular missing data mechanisms; missing completely at random (MCAR) and missing at random (MAR). We illustrate the approach with both real and simulated data.

Keywords

mediation analysisfunctional response modelsstructural equation modelweighted generalized estimating equations IImissing datageneralized least squares
Type
Original Paper
Information
Psychometrika , Volume 79 , Issue 4 , October 2014 , pp. 543 - 568
Copyright
Copyright © 2013 The Psychometric Society

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable

References

Allison, P.D. (1987). Estimation of linear models with incomplete data. In Sociological methodology (pp. 71–103). Washington: American Sociological Association.Google Scholar
Arbuckle, J.L. 1995–2010. IBM SPSS Amos 19 user’s guide. Crawfordville: Amos Development Corporation.Google Scholar
Baker, L.A., & Fulker, D.W. (1983). Incomplete covariance matrices and LISREL. Data Analyst, 1, 3–5.Google Scholar
Baron, R.M., & Kenny, D.A. (1986). The moderator–mediator variable distinction in social psychological research: concept, strategic and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182.CrossRefGoogle Scholar
Bentler, P.M. (2006). EQS 6 structural equations program manual. Encino: Multivariate Software, Inc..Google Scholar
Bollen, K.A. (1989). Structural equations with latent variables. New York: Wiley.CrossRefGoogle Scholar
Bollen, K.A., & Pearl, J., (2012). Eight myths about causality and structural equation models (Technical Report (R-393)). UCLA Cognitive Systems Laboratory, Draft chapter for S. Morgan (Ed.), Handbook of causal analysis for social research. New York: Springer.Google Scholar
Bollen, K.A., & Stine, R. (1990). Direct and indirect effects: classical and bootstrap estimates of variability. Sociological Methodology, 20, 115–140.CrossRefGoogle Scholar
Browne, M.W. (1974). Generalized least-squares estimators in the analysis of covariance structures. South African Statistical Journal, 8, 1–24.Google Scholar
Clogg, C.C., Petkova, E., & Shihadeh, E.S. (1992). Statistical methods for analyzing collapsibility in regression models. Journal of Educational Statistics, 17(1), 51–74.CrossRefGoogle Scholar
Daniels, M.J., & Kass, R.E. (2001). Shrinkage estimators for covariance matrices. Biometrics, 57, 1173–1184.CrossRefGoogle ScholarPubMed
Gunzler, D., (2011). A class of distribution-free models for longitudinal mediation analysis. Doctoral dissertation. Retrieved from ProQuest Dissertations and Theses Database (AAT 3478294).Google Scholar
Imai, K., Keele, L., & Tingley, D. (2010). A general approach to casual mediation analysis. Psychological Methods, 15(4), 309–334.CrossRefGoogle Scholar
Joreskog, K.G., & Sorbom, D. (1997). Lisrel 8 user’s guide (2nd ed.). Lincolnwood: Scientific Software.Google Scholar
Kowalski, J., & Powell, J. (2004). Nonparametric inference for stochastic linear hypotheses: application to high-dimensional data. Biometrika, 91(2), 393–408.CrossRefGoogle Scholar
Kowalski, J., & Tu, X.M. (2007). Modern applied U statistics. New York: Wiley.CrossRefGoogle Scholar
Kowalski, J., Pagano, M., & DeGruttola, V. (2002). A nonparametric test of gene region heterogeneity associated with phenotype. Journal of the American Statistical Association, 97(458), 398–408.CrossRefGoogle Scholar
Kraemer, H. (2001). How do risk factors work together? Mediators, moderators, and independent, overlapping, and proxy risk factors. The American Journal of Psychiatry, 158, 848–856.CrossRefGoogle ScholarPubMed
Little, R.J.A., & Rubin, D.B. (1987). Statistical analysis with missing data. New York: Wiley.Google Scholar
Lu, N., Tang, W., He, H., Yu, Q., Crits-Christoph, P., Zhang, H., & Tu, X.M. (2009). On the impact of parametric assumptions and robust alternatives for longitudinal data analysis. Biometrical Journal, 51, 627–643.CrossRefGoogle ScholarPubMed
Ma, Y., Tang, W., Feng, C., & Tu, X.M. (2008). Inference for kappas for longitudinal study data: applications to sexual health research. Biometrics, 64, 781–789.CrossRefGoogle ScholarPubMed
Ma, Y., Tang, W., Yu, Q., & Tu, X.M. (2010). Modeling concordance correlation coefficient for longitudinal study data. Psychometrika, 75(1), 99–119.CrossRefGoogle Scholar
Maas, C.J., & Hox, J.J. (2004). Robustness issues in multilevel regression. Statistica Neerlandica, 58, 127–137.CrossRefGoogle Scholar
MacKinnon, D. (2008). Introduction to statistical mediation analysis. New York: Lawrence Erlbaum Associates.Google Scholar
MacKinnon, D., & Fairchild, A. (2009). Current directions in mediation analysis. Current Directions in Psychological Science, 18, 16–20.CrossRefGoogle ScholarPubMed
McCullagh, P., & Nelder, J.A. (1989). Generalized linear models. London: Chapman and Hall.CrossRefGoogle Scholar
Muthén, L.K., & Muthén, B.O. 1998–2010.Mplus user’s guide (6th ed.). Los Angeles: Muthén & Muthén.Google Scholar
Nelsen, R.B. (2006). An introduction to Copulas. New York: Springer.Google Scholar
Pan, W. (2001). On the robust variance estimator in generalized estimating equations. Biometrika, 88, 901–906.CrossRefGoogle Scholar
Papadopoulos, S., & Amemiya, Y. (2005). Correlated samples with fixed and nonnormal latent variables. The Annals of Statistics, 33, 2732–2757.CrossRefGoogle Scholar
Preacher, K.J., Wichman, A.L., MacCallum, R.C., & Briggs, N.E. (2008). Latent growth curve modeling. Los Angeles: Sage.CrossRefGoogle Scholar
Prentice, R.L., & Zhao, L.P. (1991). Estimating equations for parameters in means and covariances of multivariate discrete and continuous responses. Biometrics, 47, 825–839.CrossRefGoogle ScholarPubMed
R Development Core Team (2008). R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing.Google Scholar
Reboussin, B.A., & Liang, K.Y. (1998). An estimating equations approach for the LISCOMP model. Psychometrika, 63, 165–182.CrossRefGoogle Scholar
Robins, J.M., Rotnitzky, A., & Zhao, L.P. (1995). Analysis of semiparametric regression models for repeated outcomes in the presence of missing data. Journal of the American Statistical Association, 90, 106–121.CrossRefGoogle Scholar
Rothman, K.J., & Greenland, S. (1998). Modern epidemiology. Philadelphia: Lippingcott Williams and Wilkins.Google Scholar
Satorra, A. (2002). Asymptotic robustness in multiple group linear-latent variable models. Econometric Theory, 18, 297–312.CrossRefGoogle Scholar
Satorra, A., & Bentler, P.M. (1994). Corrections to test statistics and standard errors in covariance analysis. In von Eye, A., & Clogg, C.C. (Eds.), Latent variable analysis: applications to developmental research (pp. 399–419). Newbury Park: Sage.Google Scholar
Satorra, A., & Bentler, P.M. (1998). Scaling corrections for chi-square statistics in covariance structure analysis. 1988 proceedings of the business and economics statistics section of the American Statistical Association, 308–313.Google Scholar
Scharfstein, D.O., Rotnitzky, A., & Robins, J.M. (1999). Adjusting for nonignorable drop-out using semiparametric nonresponse models. Journal of the American Statistical Association, 94, 1096–1120.CrossRefGoogle Scholar
Sobel, M.E. (1982). Asymptotic intervals for indirect effects in structural equations models. In Leinhart, S. (Eds.), Sociological methodology (pp.290–312). San Francisco: Jossey-Bass.Google Scholar
Tsiatis, A.A. (2006). Semiparametric theory and missing data. New York: Springer.Google Scholar
Tu, X.M., Feng, C., Kowalski, J., Tang, W., Wang, H., Wan, C., & Ma, Y. (2007). Correlation analysis for longitudinal data: applications to HIV and psychosocial research. Statistics in Medicine, 26, 4116–4138.CrossRefGoogle ScholarPubMed
Van der Leeden, R., & Busing, F. (1994). First iteration versus IGLS/RIGLS estimates in two-level models: a Monte Carlo study with ML3. Leiden: Department of Psychometrics and Research Methodology, Leiden University.Google Scholar
Van der Leeden, R., Busing, F., & Meijer, E., (1997). Applications of bootstrap methods for two-level models. Paper presented at the Multilevel Conference, 1–2 April, Amsterdam, The Netherlands.Google Scholar
Werts, C.E., Rock, D.A., & Grandy, J. (1979). Confirmatory factor analysis applications: missing data problems and comparisons of path models between populations. Multivariate Behavioral Research, 14, 199–213.CrossRefGoogle ScholarPubMed
Wyman, P.A., Cross, W., Brown, H.C., Yu, Q., & Tu, X.M. (2010). Intervention to strengthen emotional self-regulation in children with emerging mental health problems: proximal impact on school behavior. Journal of Abnormal Child Psychology, 30, 707–720.CrossRefGoogle Scholar
Yu, Q., Tang, W., Kowalski, J., & Tu, X.M. (2011). Multivariate U-statistics: a tutorial with applications. Wiley Interdisciplinary Reviews: Computational Statistics, 3, 457–471.CrossRefGoogle Scholar
Yu, Q., Chen, R., Tang, W., He, H., Gallop, R., Crits-Christoph, P., Hu, J., & Tu, X. M. (2013). Distribution-free models for longitudinal count responses with overdispersion and structural zeros. Statistics in Medicine, 32(14), 2390–2405.CrossRefGoogle ScholarPubMed
Zhang, Z., & Wang, L. (2013). Methods for mediation analysis with missing data. Psychometrika, 78, 154–184.CrossRefGoogle ScholarPubMed