Sensitivity Analysis of Mixed Models for Incomplete Longitudinal Data

Xu, Shu; Blozis, Shelley A.

doi:10.3102/1076998610375836

Cited by 29 publications

(29 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Strategies for dealing with data that are NMAR may rely on known analytic procedures, including mean and covariance structure analysis, but may also include a model for the missingness. Major frameworks in this area include selection models and pattern-mixture models (Demirtas & Schafer, 2003; Kenward, 1998; Little, 1995; Xu & Blozis, 2010). A major complication in the estimation of NMAR models concerns the identification of parameters that correspondence to the dependence of the missingness on the missing data (Little & Rubin, 2002).…”

Section: Discussionmentioning

confidence: 99%

Sensitivity Analysis of Multiple Informant Models When Data Are Not Missing at Random

Blozis

et al. 2013

Structural Equation Modeling: A Multidisciplinary Journal

Self Cite

View full text Add to dashboard Cite

Missing data are common in studies that rely on multiple informant data to evaluate relationships among variables for distinguishable individuals clustered within groups. Estimation of structural equation models using raw data allows for incomplete data, and so all groups may be retained even if only one member of a group contributes data. Statistical inference is based on the assumption that data are missing completely at random or missing at random. Importantly, whether or not data are missing is assumed to be independent of the missing data. A saturated correlates model that incorporates correlates of the missingness or the missing data into an analysis and multiple imputation that may also use such correlates offer advantages over the standard implementation of SEM when data are not missing at random because these approaches may result in a data analysis problem for which the missingness is ignorable. This paper considers these approaches in an analysis of family data to assess the sensitivity of parameter estimates to assumptions about missing data, a strategy that may be easily implemented using SEM software.

show abstract

Section: Discussionmentioning

confidence: 99%

Sensitivity Analysis of Multiple Informant Models When Data Are Not Missing at Random

Blozis

et al. 2013

Structural Equation Modeling: A Multidisciplinary Journal

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, LMM, a likelihood-based method, is valid under the missing at random (MAR) assumption that missingness is unrelated to any unobserved variable. Although verifying the MAR assumption with observed data may not be possible, this assumption is commonly used in the literature (Xu & Blozis, 2011). Statistical Analysis Software was used to perform the analyses.…”

Section: Methodsmentioning

confidence: 99%

Does treatment readiness enhance the response of African American substance users to Motivational Enhancement Therapy?

Burlew

Montgomery²,

Kosinski³

et al. 2013

Psychology of Addictive Behaviors

View full text Add to dashboard Cite

The development of effective treatments for African Americans and other ethnic minorities is essential for reducing health disparities in substance use. Despite research suggesting that Motivational Enhancement Therapy (MET) may reduce substance use among African Americans, the findings have been inconsistent. This research examined the extent to which readiness-to-change (RTC) affects response to MET among African American substance users. The study was a secondary analysis of the 194 African American substance users participating in a multisite randomized clinical trial evaluating MET originally conducted within the National Drug Abuse Treatment Clinical Trials Network. The participants were randomly assigned to receive either three sessions of MET or Counseling-As-Usual (CAU) followed by the ordinary treatment and other services offered at the five participating outpatient programs. Participants were categorized as either high or lower on RTC based on their scores on the University of Rhode Island Change Assessment. The participants reported their substance use at baseline and throughout the 16 weeks after randomization. Among the high RTC participants, those in MET tended to report fewer days of substance use per week over time than participants in CAU. However, among the lower RTC participants, the CAU group tended to report fewer days of substance use over time than MET participants. In contrast to previous thinking, the findings suggest that MET may be more effective for high than lower RTC African American participants.

show abstract

“…We use various structures that are available within our multi-response model to perform a sensitivity analysis (Xu and Blozis, 2011) on the teacher rankings produced when analyzing a data set containing semester calculus grades from a large public university. We find that the rankings of teacher effects may change depending on the assumptions made about the structure of the missing data mechanism.…”

Section: Department Of Education 2009) Given the Magnitude Of The Dmentioning

confidence: 99%

“…If any of the MNAR models were to produce substantially different results from the standard GP model, this would indicate that the conclusions of the VAM depend on assumptions made about the nature of the missing observations. However, the appropriate missing data process cannot be chosen by empirical investigation of the observed data (including examination of the log-likelihood) since the observed data do not provide information to support one particular MNAR model over another (Fitzmaurice et al, 2004;Xu and Blozis, 2011). As stated by Molenberghs and Kenward (2007), "ignoring MNAR models is no different an option than shifting to one particular MNAR model, it is just much more convenient."…”

Section: Sensitivity Analysismentioning

confidence: 99%

A Correlated Random Effects Model for Nonignorable Missing Data in Value-Added Assessment of Teacher Effects

Karl¹,

Yang

Lohr

2013

Journal of Educational and Behavioral Statistics

View full text Add to dashboard Cite

Value-added models (VAMs) are used by many states to assess contributions of individual teachers and schools to students' academic growth. The generalized persistence VAM, one of the most flexible in the literature, estimates the "value added" by individual teachers to their students' current and future test scores by employing a mixed model with a longitudinal database of test scores. There is concern, however, that missing values that are common in the longitudinal student scores can bias value-added assessments, especially when the models serve as a basis for personnel decisions -such as promoting or dismissing teachers -as they are being used in some states. Certain types of missing data require that the VAM be modeled jointly with the missingness process in order to obtain unbiased parameter estimates.This dissertation studies two problems. First, the flexibility and multimembership random effects structure of the generalized persistence model lead to computational challenges that have limited the model's availability. To this point, no methods have been developed for scalable maximum likelihood estimation of the model. An EM algorithm to compute maximum likelihood estimates efficiently is developed, making use of the sparse structure of the random effects and error covariance matrices. The algorithm is implemented in the package GPvam in R statistical software. Illustrations of the gains in computational efficiency achieved by the estimation procedure are given.Furthermore, to address the presence of potentially nonignorable missing data, a flexible correlated random effects model is developed that extends the generalized persistence model to jointly model the test scores and the missingness process, allowing the process to depend on both students and teachers. The joint model gives the ability to test the sensitivity of the VAM to the presence of noni ignorable missing data. Estimation of the model is challenging due to the nonhierarchical dependence structure and the resulting intractable high-dimensional integrals. Maximum likelihood estimation of the model is performed using an EM algorithm with fully exponential Laplace approximations for the E step. The methods are illustrated with data from university calculus classes and with data from standardized test scores from an urban school district.ii To Laura, Oliver, and the rest of my family.iii ACKNOWLEDGEMENTS I owe a heartfelt thanks to my advisors, Dr. Lohr and Dr. Yang, who have dedicated so much of their time and effort to helping me develop this dissertation.They were selfless in their willingness to respond quickly to all of my emails and to meet with me whenever I needed their assistance. I am indebted to them for significant portions of my graduate education that they went out of their way to provide me with, including encouragement and funding to travel to JSM in Vancouver and Miami Beach, help with job searching, guiding me through the publication and peer-review process, and in general teaching me how to work in an academic environme...

show abstract

Sensitivity Analysis of Mixed Models for Incomplete Longitudinal Data

Cited by 29 publications

References 31 publications

Sensitivity Analysis of Multiple Informant Models When Data Are Not Missing at Random

Sensitivity Analysis of Multiple Informant Models When Data Are Not Missing at Random

Does treatment readiness enhance the response of African American substance users to Motivational Enhancement Therapy?

A Correlated Random Effects Model for Nonignorable Missing Data in Value-Added Assessment of Teacher Effects

Contact Info

Product

Resources

About