Abstract
Psychologists often use latent transition analysis (LTA) to investigate state-to-state change in discrete latent constructs involving delinquent or risky behaviors. In this setting, latent-state-dependent nonignorable missingness is a potential concern. For some longitudinal models (e.g., growth models), a large literature has addressed extensions to accommodate nonignorable missingness. In contrast, little research has addressed how to extend the LTA to accommodate nonignorable missingness. Here we present a shared parameter LTA that can reduce bias due to latent-state-dependent nonignorable missingness: a parallel-process missing-not-at-random (MNAR-PP) LTA. The MNAR-PP LTA allows outcome process parameters to be interpreted as in the conventional LTA, which facilitates sensitivity analyses assessing changes in estimates between LTA and MNAR-PP LTA. In a sensitivity analysis for our empirical example, previous and current membership in high-delinquency states predicted adolescents’ membership in missingness states that had high nonresponse probabilities for some or all items. A conventional LTA overestimated the proportion of adolescents ending up in a low-delinquency state, compared to an MNAR-PP LTA.
Similar content being viewed by others
Notes
The LTA model is also known as a multiple-indicator latent Markov model (e.g., Langeheine & Van de Pol, 1990).
The \(Q=1\) state model necessarily has a different structure; it simply consists of estimating a set of missingness indicator thresholds for that single class.
Suppose the MNAR-SP LTA results indicated that a particular state at time \(t\) evidenced a high probability of violent behavior and a high probability of missingness. Persons might obtain a similarly high posterior probability of assignment to that MNAR-SP LTA state simply by endorsing many aggressive items (even if they had no missingness), or by having many missing responses (even if they endorsed no aggressive items). Yet, substantively, researchers would want to be able to distinguish among such persons.
References
Albert, P. (2000). A transitional model for longitudinal binary data subject to nonignorable missing data. Biometrics, 56, 602–608.
Albert, P., & Follmann, D. (2003). A random effects transition model for longitudinal binary data with informative missingness. Statistica Neerlandica, 57, 100–111.
Akaike, H. (1974). A new look at the statistical model identification. IEEE Transactions on Automatic Control, 19, 716–723.
Baggio, S., Studer, J., Deline, N’ Goran, Dupuis, M., Henchoz, Y., Mohler-Kuo, M., et al. (2014). Patterns and transitions in substance use among young Swiss men: A latent transition analysis approach. Journal of Drug Issues. doi:10.1177/0022042614526996.
Bair-Merritt, M., Ghazarian, S., Burrell, L., & Duggan, A. (2012). Patterns of intimate partner violence in mothers at-risk for child maltreatment. Journal of Family Violence, 27, 287–294.
Beunckens, C., Molenberghs, G., Thijs, H., & Verbeke, G. (2007). Incomplete hierarchical data. Statistical Methods in Medical Research, 16, 457–492.
Bray, B., Lanza, S., & Collins, L. (2010). Modeling relations among discrete developmental processes: A general approach to associative latent transition analysis. Structural Equation Modeling, 17, 541–569.
Cain, A., Epler, A., Steinley, D., & Sher, K. (2012). Concerns related to eating, weight, and shape: Typologies and transitions in men during the college years. International Journal of Eating Disorders, 45, 768–775.
Carpenter, J., Pocock, S., & Lamm, C. (2002). Coping with missing values in clinical trials: A model based approach applied to asthma trials. Statistics in Medicine, 21, 1043–1066.
Catts, H., Compton, D., Tomblin, J., & Bridges, M. (2012). Prevalence and nature of late-emerging poor readers. Journal of Educational Psychology, 104, 166–181.
Chen, B., Yi, G., & Cook, R. (2011). Progressive multi-state models for informatively incomplete longitudinal data. Journal of Statistical Planning and Inference, 141, 80–93.
Cleveland, M., Lanza, S., Ray, A., Turrisi, R., & Mallett, K. (2012). Transitions in first-year college student drinking behaviors: Does pre-college drinking moderate the effects of parent- and peer-based intervention components? Psychology of Addictive Behaviors, 26, 440–450.
Cochran, G., Field, C., & Caetano, R. (2013). Changes in classes of injury-related risks and consequences of risk-level drinking: A latent transition analysis. Journal of Behavioral Health Services & Research,. doi:10.1007/s11414-013-9378-3.
Collins, L., & Lanza, S. (2010). Latent class and latent transition analysis. Hoboken, NJ: Wiley.
Collins, L., & Wugalter, S. (1992). Latent class models for stage-sequential dynamic latent variables. Multivariate Behavioral Research, 27, 131–157.
Cole, B., Bonetti, M., Zaslavsky, A., & Gelber, R. (2005). A multistate Markov chain model for longitudinal, categorical quality-of-life data subject to non-ignorable missingness. Statistics in Medicine, 24, 2317–2334.
Cook, E., Pflieger, J., Connell, A., & Connell, C. (2014). Do specific transitional patterns of antisocial behavior during adolescence increase risk for problems in young adulthood? Journal of Abnormal Child Psychology. doi:10.1007/s10802-014-9880-y.
Courtney, M., & Cusick, G. (2007). Crime during the transition to adulthood: How youth fare as they leave out-of-home care in Illinois, Iowa and Wisconsin. [Computer file].
Dempster, A., Laird, N., & Rubin, D. (1977). Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society, Series B, 39, 1–38.
Diggle, P., & Kenward, M. G. (1994). Informative drop-out in longitudinal data-analysis. Applied Statistics C, 43, 49–93.
Enders, C. K. (2010). Applied missing data analysis. New York, NY: Guilford.
Enders, C. K. (2011). Missing not at random models for latent growth curve analyses. Psychological Methods, 16, 1–16.
Feldman, B., & Rabe-Hesketh, S. (2012). Modeling achievement trajectories when attrition is informative. Journal of Educational and Behavioral Statistics, 37, 703–736.
Flaherty, B. P. (2008). Testing the degree of cross-sectional and longitudinal dependence between two discrete dynamic processes. Developmental Psychology, 44, 468–480.
Gottfredson, N., Bauer, D., & Baldwin, S. (2014). Modeling change in the presence of nonrandomly missing data: Evaluating a shared parameter mixture model. Structural Equation Modeling, 21, 196–209.
Groves, R., Dillman, D., Eltinge, J., & Little, R. (2002). Survey nonresponse. New York: Wiley.
Haviland, A., Jones, B., & Nagin, D. (2011). Group-based trajectory modeling extended to account for nonrandom participant attrition. Sociological Methods & Research, 40, 367–390.
Hedeker, D., & Gibbons, R. D. (1997). Application of random-effects pattern-mixture models for missing data in longitudinal studies. Psychological Methods, 2, 64–78.
Hinshaw, S., Lahey, B., & Hart, E. (1993). Issues of taxonomy and comorbidity in the development of conduct disorder. Development and Psychopathology, 5, 31–49.
Hopfer, S., Hecht, M., Lanza, S., Tan, X., & Xu, S. (2013). Preadolescent drug use resistance skill profiles substance use, and substance use prevention. Journal of Primary Prevention, 34, 395–404.
Ibrahim, J. G., Chen, M.-H., Lipsitz, S., & Herring, A. (2005). Missing-data methods for generalized linear models: A comparative review. Journal of the American Statistical Association, 100, 332–346.
Jackson, K., Sher, K., & Schulenberg, J. (2005). Conjoint developmental trajectories of young adult alcohol and tobacco use. Journal of Abnormal Psychology, 114, 612–626.
Jansen, I., Hens, N., Molenberghs, G., et al. (2006). The nature of sensitivity in missing not at random models. Computational Statistics & Data Analysis, 50, 830–858.
Kroesen, M. (2014). Modeling the behavioral determinants of travel behavior: An application of latent transition analysis. Transportation Research, 65, 56–67.
Kurland, B., & Heagerty, P. (2004). Marginalized transition models for longitudinal binary data with ignorable and non-ignorable drop-out. Statistics in Medicine, 23, 2673–2695.
Lahey, B., & Loeber, R. (1994). Framework for a developmental model of oppositional defiant disorder and conduct disorder. In D. K. Routh (Ed.), Disruptive behavior disorders in childhood. New York: Plenum.
Langeneine, R., & van de Pol, F. (1990). A unifying framework for Markov modeling in discrete space and discrete time. Sociological Methods and Research, 18, 416–441.
Lanza, S., & Collins, L. (2008). A new SAS procedure for latent transition analysis: Transitions in dating and sexual risk behavior. Developmental Psychology, 44, 446–456.
Lee, S., & Song, X.-Y. (2003). Maximum likelihood estimation and model comparison for mixtures of structural equation models with ignorable missing data. Journal of Classification, 20, 221–255.
Lee, B., & Vondracek, F. (2014). Teenage goals and self-efficacy beliefs as precursors of adult career and family outcomes. Journal of Vocational Behavior, 85, 228–237.
Lin, H., Liu, D., & Zhou, X. (2010). A correlated random-effects model for normal longitudinal data with nonignorable missingness. Statistics in Medicine, 29, 236–247.
Little, R. J. A. (1995). Modeling the drop-out mechanism in repeated-measures studies. Journal of the American Statistical Association, 90, 1112–1121.
Little, R. J. A. (2008). Selection and pattern-mixture models. In G. Fitzmaurice, et al. (Eds.), Advances in longitudinal data analysis (pp. 409–431). London: CRC Press.
Little, R.J.A., D’Agostino, R., Cohen, M., Dickerson, K., Emerson, S., & Farrar, J. et al. (2012). The prevention and treatment of missing data in clinical trials: Special report. The New England Journal of Medicine, 1355–1360.
Little, R. J. A., & Rubin, D. B. (1999). Adjusting for nonignorable drop-out using semiparametric nonresponse models: Comment. Journal of the American Statistical Association, 94, 1130–1132.
Little, R. J. A., & Rubin, D. B. (2002). Statistical analysis with missing data. Hoboken, NJ: Wiley.
Liu, X., Waternaux, C., & Petkova, E. (1999). Influence of human immunodeficiency virus infection on neurological impairment: An analysis of longitudinal binary data with informative drop-out. Applied Statistics, 48, 103–115.
Loeber, R., Keenan, K., & Zhang, Q. (1997). Boys’ experimentation and persistence in developmental pathways toward serious delinquency. Journal of Child and Family Studies, 6, 321–357.
Lu, Z. L., Zhang, Z., & Lubke, G. (2011). Bayesian inference for growth mixture models with latent class dependent missing data. Multivariate Behavioral Research, 46, 567–597.
Lu, Z. L., & Zhang, Z. (2014). Robust growth mixture models with non-ignorable missingness: Models, estimation, selection, and application. Computational Statistics & Data Analysis, 71, 220–240.
Mackesy-Amiti, M., Ouellet, L., Finnegan, L., Hagan, H., Golub, E., Latka, M., et al. (2014). Transitions in latent classes of sexual risk behavior among young injection drug users following HIV prevention intervention. AIDS Behavior, 18, 464–472.
Mallinckrodt, C., Sanger, T., Dube, S., DeBrota, D., Molenberghs, G., Carroll, R., et al. (2003). Assessing and interpreting treatment effects in longitudinal clinical trials with missing data. Biological Psychiatry, 53, 754–760.
McLachlan, G., & Peel, D. (2000). Finite mixture models. New York, NY: Wiley.
McMillen, J., Zima, B., Scott, L., et al. (2005). Prevalence of psychiatric disorders among older youths in the foster care system. Journal of the American Academy of Child and Adolescent Psychiatry, 44, 88–95.
Meier, M., Caspi, A., Houts, R., Slutske, W., Harrington, H., Jackson, K., et al. (2013). Prospective developmental subtypes of alcohol dependence from age 18 to 32 years: Implications for nosology, etiology, and intervention. Development and Psychopathology, 25, 785–800.
Minini, P., & Chavance, M. (2003). Sensitivity analysis of longitudinal normal data. Statistics in Medicine, 23, 1039–1054.
Moffitt, T. E., Arseneault, L., Jaffee, S., et al. (2008). DSM-V conduct disorder: Research needs for an evidence base. Journal of Child Psychology & Psychiatry, 49, 3–33.
Molenberghs, G., & Kenward, M. G. (2007). Missing data in clinical studies. Chichester: Wiley.
Molenberghs, G., & Verbeke, G. (2005). Models for discrete longitudinal data. New York: Springer.
Muthén, B., Asparouhov, T., Hunter, A. M., & Leuchter, A. F. (2011). Growth modeling with nonignorable dropout: Alternative analyses of the STAR*D antidepressant trial. Psychological Methods, 16, 17–33.
Muthén, L.K., Muthén, B.O. (1998–2014). Mplus 7 users guide. Los Angeles, CA: Muthén & Muthén.
Nagin, D. S. (2005). Group-based modeling of development. Cambridge, MA: Harvard Press.
Nagin, D. S., & Tremblay, R. E. (2001). Analyzing developmental trajectories of distinct but related behaviors: A group-based method. Psychological Methods, 6, 18–34.
National Research Council (2010). The prevention and treatment of missing data in clinical trials. In Panel on handling missing data in clinical trials. Division of Behavioral and Social Sciences and Education. Washington, D.C.: National Academies Press.
Nylund, K., Asparouhov, T., & Muthén, B. (2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte-Carlo simulation study. Structural Equation Modeling, 14, 535–569.
Nylund, K., Muthén, B. Nishina, A., Bellmore, A., Graham, S. (2006). Stability and instability of peer victimization during middle school: Using latent transition analysis with covariates, distal outcomes, and modeling extensions. www.statmodel.com.
Raftery, A. (1995). Bayesian model selection in social research. Sociological Methodology, 25, 111–163.
Reboussin, B. A., Reboussin, D. M., Liang, K., & Anthony, J. C. (1998). Latent transition modeling of progression of health risk behaviors. Multivariate Behavioral Research, 33, 457–478.
Rizopoulos, D., Verbeke, G., Lesaffre, E., & Vanrenterghem, Y. (2008). A two-part joint model for the analysis of survival and longitudinal binary data with excess zeros. Biometrics, 64, 611–619.
Rodgers, S., Ajdacic-Gross, V., Muller, M., Hengartner, M., Holtforth, M., Angst, J., et al. (2013). The role of sex on stability and change of depression symptom subtypes over 20 years: A latent transition analysis. European Archives of Psychiatry: DOI. doi:10.1007/s00406-013-0475-3.
Rotnitzky, A., Robins, J., & Scharfstein, D. (1998). Semiparametric regression for repeated outcomes with nonignorable nonresponse. Journal of the American Statistical Association, 93, 1321–1339.
Roy, J. (2003). Modeling longitudinal data with nonignorable dropouts using a latent dropout class model. Biometrics, 59, 829–836.
Rubin, D. B. (1976). Inference and missing data. Biometrika, 63, 581–592.
Schafer, J. L., & Graham, J. W. (2002). Missing data: Our view of the state of the art. Psychological Methods, 7, 147–177.
Schwarz, G. E. (1978). Estimating the dimension of a model. Annals of Statistics, 6, 461–464.
Soto-Ramirez, N., Ziyab, A., Karmaus, W., Zhang, H., Kurukulaaratchy, R., Ewart, S., et al. (2013). Epidemiologic methods of assessing asthma and wheezing episodes in longitudinal studies: Measures of change and stability. Journal of Epidemiology, 23, 399–410.
Sterba, S. K. (2013). Understanding linkages among mixture models. Multivariate Behavioral Research, 48, 775–815.
Sterba, S.K. & Gottfredson, N. (2014). Diagnosing global case influence on MAR versus MNAR model comparisons. Structural Equation Modeling, In press.
Sterba, S. K., & Pek, J. (2012). Individual influence on model selection. Psychological Methods, 17, 582–599.
Ten Have, T., Kunselman, A., Pulkstenis, E., & Landis, J. (1998). Mixed effects logistic regression models for longitudinal binary response data with informative drop-out. Biometrics, 54, 367–383.
Thijs, H., Molenberghs, G., Michiels, B., Verbeke, G., & Curran, D. (2002). Strategies to fit pattern-mixture models. Biostatistics, 3, 245–265.
Tofighi, D., & Enders, C. (2007). Identifying the correct number of classes in growth mixture models. In G. R. Hancock & K. M. Samuelsen (Eds.), Advances in latent variable mixture models (pp. 317–341). Greenwich, CT: Information Age.
Velicer, W., Martin, R., & Collins, L. (1996). Latent transition analysis for longitudinal data. Addiction, 91, S197–S209.
Verbeke, G., Lessafre, E., & Spiessens, B. (2001). The practical use of different strategies to handle dropout in longitudinal studies. Drug Information Journal, 35, 419–434.
Williford, A., Boulton, A., & Jenson, J. (2014). Transitions between subclasses of bullying and victimization when entering middle school. Aggressive Behavior, 40, 24–41.
Witkiewitz, K., & Villarroel, N. (2009). Dynamic association between negative affect and alcohol lapses following alcohol treatment. Journal of Consulting and Clinical Psychology, 77, 633–644.
Witkiewitz, K. (2008). Lapses following alcohol treatment: Modeling the falls from the wagon. Journal of Studies on Alcohol and Drugs, 69, 594–604.
White, T., & Erosheva, E. (2013). Using group-based latent class transition models to analyze chronic disability data from the National Long-Term Care Survey 1984–2004. Statistics in Medicine, 32, 3569–3589.
Wu, M., & Carroll, R. (1988). Estimation and comparison of changes in the presence of informative right censoring by modeling the censoring process. Biometrics, 44, 175–188.
Xu, S., & Blozis, S. (2011). Sensitivity analyses of mixed models for incomplete longitudinal data. Journal of Educational and Behavioral Statistics, 36, 237–256.
Yang, M., & Maxwell, S. (2014). Treatment effects in randomized longitudinal trials with different types of nonignorable dropout. Psychological Methods, 19, 188–210.
Yang, X., Shoptaw, S., Nie, K., Liu, J., & Belin, T. (2007). Markov transition models for binary repeated measures with ignorable and nonignorable missing values. Statistical Methods in Medical Research, 16, 347–364.
Zhang, Z., & Wang, L. (2012). A note on the robustness of a full Bayesian method for nonignorable missing data analysis. Brazilian Journal of Probability and Statistics, 26, 244–264.
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Sterba, S.K. A Latent Transition Analysis Model for Latent-State-Dependent Nonignorable Missingness. Psychometrika 81, 506–534 (2016). https://doi.org/10.1007/s11336-015-9442-4
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11336-015-9442-4