Often in educational and behavioral randomized trials, researchers discover that subgroups of individuals respond differently to the same intervention due to underlying individual differences. Previous studies have investigated such heterogeneous treatment effects (HTE) across different latent classes for both cross-sectional and longitudinal (repeated measures) outcomes, but not in the context of longitudinal mediation analysis. This study develops Bayesian (nonlinear) growth mixture mediation models, B(N)GMMMs, to assess the HTE of the intervention variable X on the longitudinal dependent variable Y, via the longitudinal mediator M. We consider Y following linear or nonlinear trajectories and incorporate class predictive and growth predictive covariates. Monte Carlo simulations were conducted to evaluate model performance, and an empirical example demonstrates the model’s application.
BatesD. M.WattsD. J. (1988). Nonlinear regression: Iterative estimation and linear approximations. In Nonlinear regression analysis and its applications (pp. 3266). John Wiley & Sons. https://doi.org/10.1002/9780470316757.ch2
2.
BibriescasN.WhittakerT. A. (2023). Detecting latent classes through mediation in regression mixture models. Structural Equation Modeling: A Multidisciplinary Journal, 30(3), 449–457.
3.
BlozisS. A.CudeckR. (1999). Conditionally linear mixed-effects models with latent variable covariates. Journal of Educational and Behavioral Statistics, 24(3), 245–270.
4.
CheongJ.MacKinnonD. P.KhooS. T. (2003). Investigation of mediational processes using parallel process latent growth curve modeling. Structural Equation Modeling: A Multidisciplinary Journal, 10(2), 238. https://doi.org/10.1207/S15328007SEM1002_5
5.
ChoiS.McMasterK. L.KohliN.ShanahanE.BirinciS.AnJ.Duesenberg-MarshallM.LembkeE. S. (2024). Longitudinal effects of data-based instructional changes for students with intensive learning needs: A piecewise linear–linear mixed-effects modeling approach. Journal of Educational Psychology, 116(4), 608.
6.
CohenJ. (1988). Statistical power analysis for the behavioral sciences, 2nd ed.Hillsdale, NJ: Erlbaum.
7.
CrowderM.HandD. (1990). Analysis of repeated measures, London: Chapman and Hall.
8.
DepaoliS.BoyajianJ. (2014). Linear and nonlinear growth models: Describing a Bayesian perspective. Journal of consulting and clinical psychology, 82(5), 784.
9.
FairchildA. J.MacKinnonD. P.TaborgaM. P.TaylorA. B. (2009). R2 effect-size measures for mediation analysis. Behavior Research Methods, 41(2), 486–498.
10.
HarwellM. (2019). A strategy for using bias and RMSE as outcomes in Monte Carlo studies in statistics. Journal of Modern Applied Statistical Methods, 17(2), P2938. https://doi.org/10.22237/jmasm/1551907966
11.
KohliN.HarringJ. R.ZopluogluC. (2016). A finite mixture of nonlinear random coefficient models for continuous repeated measures data. Psychometrika, 81(3), 851–880.
12.
KohliN.HughesJ.WangC.ZopluogluC.DavisonM. L. (2015). Fitting a linear–linear piecewise growth mixture model with unknown knots: A comparison of two common approaches to inference. Psychological Methods, 20(2), 259–275. https://doi.org/10.1037/met0000034
13.
KohliN.SullivanA. L.SadehS.ZopluogluC. (2015). Longitudinal mathematics development of students with learning disabilities and students without disabilities: A comparison of linear, quadratic, and piecewise linear mixed effects models. Journal of School Psychology, 53(2), 105–120.
14.
LachowiczM. J.PreacherK. J.KelleyK. (2018). A novel measure of effect size for mediation analysis. Psychological Methods, 23(2), 244.
15.
LakensD. (2013). Calculating and reporting effect sizes to facilitate cumulative science: A practical primer for t-tests and ANOVAs. Frontiers in Psychology, 4, Article 863. https://doi.org/10.3389/fpsyg.2013.00863
16.
LammR. Z. (2022). Incorporation of covariates in Bayesian piecewise growth mixture models [Doctoral dissertation, University of Minnesota].
17.
LairdN. M.WareJ. H. (1982). Random-effects models for longitudinal data. Biometrics, 38(4), 963–974. https://doi.org/10.2307/2529876
18.
LiuJ.PereraR. A. (2023). Extending mixture of experts model to investigate heterogeneity of trajectories: When, where, and how to add which covariates. Psychological Methods, 28(1), 152.
19.
LiuX.ValentinoK.WangL. (2022). The impact of omitting confounders in laten growth curve mediation modeling: Analytical examination and methods for sensitivity analysis. Multivariate Behavioral Research, 57, 153–154. https://doi.org/10.1080/00273171.2021.2011701
20.
LiuX.ZhangZ.ValentinoK.WangL. (2024). The impact of omitting confounders in parallel process latent growth curve mediation models: Three sensitivity analysis approaches. Structural Equation Modeling: A Multidisciplinary Journal, 31(1), 132–150. https://doi.org/10.1080/10705511.2023.2189551
21.
LockE. F.KohliN.BoseM. (2018). Detecting multiple random changepoints in Bayesian piecewise growth mixture models. Psychometrika, 83(3), 733–750. https://doi.org/10.1007/s11336-017-9594-5
22.
MiočevićM.O’RourkeH. P.MacKinnonD. P.BrownH. C. (2018). Statistical properties of four effect-size measures for mediation models. Behavior Research Methods, 50, 285–301.
23.
MuthénB.BrownC. H.MasynK.JoB.KhooS. T.YangC. C.WangC. P.KellamS. G.CarlinJ. B.LiaoJ. (2002). General growth mixture modeling for randomized preventive interventions. Biostatistics, 3(4), 459–475.
24.
NeelonB.SwamyG. K.BurgetteL. F.MirandaM. L. (2011). A Bayesian growth mixture model to examine maternal hypertension and birth outcomes. Statistics in medicine, 30(22), 2721–2735.
25.
PeraltaY.KohliN.LockE. F.DavisonM. L. (2022). Bayesian modeling of associations in bivariate piecewise linear mixed-effects models. Psychological Methods, 27(1), 44–64. https://doi.org/10.1037/met0000358
26.
PetrowskiK.SchmalbachB.JaglaM.FrankeG. H.BrählerE. (2018). Norm values and psychometric properties of the brief symptom inventory-18 regarding individuals between the ages of 60 and 95. BMC medical research methodology, 18(1), 164.
R Core Team. (2023). R: A language and environment for statistical computing. R Foundation for Statistical Computing. https://www.R-project.org/
29.
RoeschS. C.NormanG. J.AdamsM. A.KerrJ.SallisJ. F.RyanS.CalfasK. J.PatrickK. (2009). Latent growth curve modeling of adolescent physical activity: Testing parallel process and mediation models. Journal of Health Psychology, 14(2), 313–325.
30.
RohloffC. T.KohliN.LockE. F. (2024). Identifiability and estimability of Bayesian linear and nonlinear crossed random effects models. British Journal of Mathematical and Statistical Psychology, 77(2), 375–394.
31.
RosserB. R. S. (2024). Restore 2 [Data set]. Inter-university Consortium for Political and Social Research. https://doi.org/10.3886/E202162V1
32.
SerangS.ZhangZ.HelmJ.SteeleJ. S.GrimmK. J. (2015). Evaluation of a Bayesian approach to estimating nonlinear mixed-effects mixture models. Structural Equation Modeling: A Multidisciplinary Journal, 22(2), 202–215.
33.
TurneyK. (2015). Beyond average effects: Incorporating heterogeneous treatment effects into family research. Journal of Family Theory & Review, 7(4), 468–481.
34.
WeiJ. T.DunnR. L.LitwinM. S.SandlerH. M.SandaM. G. (2000). Development and validation of the expanded prostate cancer index composite (EPIC) for comprehensive assessment of health-related quality of life in men with prostate cancer. Urology, 56(6), 899–905.
35.
WillkeR. J.ZhengZ.SubediP.AlthinR.MullinsC. D. (2012). From concepts, theory, and evidence of heterogeneity of treatment effects to methodological approaches: A primer. BMC Medical Research Methodology, 12, 185.
36.
ZhangZ.KohliN.LockE. F. (2024). Bayesian (non)linear random effects mediation models: Evaluating the impact of omitting confounders. Psychological Methods. DOI: 10.1037/met0000721
37.
ZhangZ.RohloffC. T.KohliN. (2023). Model fit indices for random effects models: Translating model fit ideas from latent growth curve models. Structural Equation Modeling: A Multidisciplinary Journal, 30(5), 822–830.
