The impact of ignoring multiple membership data structures in multilevel models

Chung, Hyewon; Beretvas, S. Natasha

doi:10.1111/j.2044-8317.2011.02023.x

Cited by 73 publications

(119 citation statements)

References 25 publications

Supporting

Mentioning

104

Contrasting

Order By: Relevance

“…In addition, rates similar to these have been used in other educational mobility simulation studies (Chung & Beretvas, 2012;Grady, 2010;Luo & Kwok, 2012). Last, the low mobility rate value is similar to the mobility rate found in the real data analysis (11.4%).…”

Section: Mobility Ratesupporting

confidence: 58%

“…Differences between multiple membership and typical multilevel models that ignore the multiple membership structure for baseline unconditional fixed effects estimates have not frequently been found in previous research (Chung & Beretvas, 2012;Grady, 2010;Luo & Kwok, 2009;Luo & Kwok, 2012;Meyers & Beretvas, 2006). However, in a study by Smith (2012), substantial positive relative parameter bias was found when estimating the level-1 predictor for the MMREM and HLM models, which was attributed to model misspecification.…”

Section: Baseline Unconditional Fixed Effectsmentioning

confidence: 82%

“…For policymakers and educational interventionists, it is also important for them to be aware that student growth analyses based on a model that does not handle mobility would model the students' growth as only being affected by one school, rather than all schools attended. Evaluating school-or student-level characteristics with a misspecified model impacts the validity of resulting statistical inferences and associated conclusions (Chung & Beretvas, 2012;Luo & Kwok, 2009;Luo & Kwok, 2012;Meyers & Beretvas, 2006). In addition, the results from this study are applicable not only to educational research, but to any applied research entailing a three-level data structure with mobility of individuals across clusters.…”

Section: List Ofmentioning

confidence: 91%

“…Previous research has demonstrated that incorrect model specification in the presence of student mobility can negatively impact parameter estimates. Previous simulation studies have shown that model misspecification can lead to inaccurate estimates of between-schools variance components and standard errors of the fixed effects (Chung & Beretvas, 2012;Grady, 2010;Luo & Kwok, 2009;Luo & Kwok, 2012;Meyers & Beretvas, 2006). The purpose of this research is then to investigate the estimation of a three-level latent variable regression model with crossclassified multiple membership data, resulting in the utilization of a CCMM-LVR model.…”

Section: List Ofmentioning

confidence: 99%

See 3 more Smart Citations

Estimating a Three-Level Latent Variable Regression Model With Cross-Classified Multiple Membership Data

Leroux

Beretvas

2018

Methodology

Self Cite

View full text Add to dashboard Cite

The current study proposed a new model, termed the cross-classified multiple membership latent variable regression (CCMM-LVR) model, to be utilized for multiple membership data structures (for example, in the presence of student mobility across schools) that provides an extension to the three-level latent variable regression model (HM3-LVR). The HM3-LVR model is beneficial for testing more flexible, directional hypotheses about growth trajectory parameters and handles pure clustering of participants within higher-level units. However, the HM3-LVR model involves the assumption that students remain in the same cluster (school) throughout the duration of the time period of interest. The CCMM-LVR model, on the other hand, appropriately models the participants' changing clusters over time.The first purpose of this study was to demonstrate use and interpretation of the CCMM-LVR model and its parameters with a large-scale longitudinal dataset that had a multiple membership data structure (i.e., student mobility). The impact of ignoring v mobility in the real data was investigated by comparing parameter estimates, standard error estimates, and model fit indices for the two estimating models (CCMM-LVR and HM3-LVR). The second purpose of the dissertation was to conduct a simulation study to try to understand the source of potential differences between the two estimating models and find out which model's estimates were closer to the truth given the conditions investigated. The manipulated conditions in the simulation study included the mobility rate, number of clustering units, number of individuals (i.e., students) per cluster (here, school), and number of measurement occasions per individual. The outcomes investigated in the simulation study included relative parameter bias, relative standard error bias, root mean square error, and coverage rates of the 95% credible intervals.Substantial bias was found across conditions for both models, but the CCMM-LVR model resulted in the least amount of relative parameter bias and more efficient estimates of the parameters, especially for larger numbers of clustering units. The results of the real data and simulation studies are discussed, along with the implications for applied researchers for when to consider using the CCMM-LVR model versus the misspecified HM3-LVR model. vi

show abstract

Section: Mobility Ratesupporting

confidence: 58%

Section: Baseline Unconditional Fixed Effectsmentioning

confidence: 82%

Section: List Ofmentioning

confidence: 91%

Section: List Ofmentioning

confidence: 99%

See 2 more Smart Citations

Estimating a Three-Level Latent Variable Regression Model With Cross-Classified Multiple Membership Data

Leroux

Beretvas

2018

Methodology

Self Cite

View full text Add to dashboard Cite

show abstract

“…As for now, optimal design recommendations for such trials are lacking. Another example is multiple membership models, see Chung and Beretvas (2012), Luo and Kwok (2012) for the implications of ignoring multiple membership structures, and Roberts and Walwyn (2013) for optimal design methodology. Thus far, the main focus has been on linear models with continuous outcome scores; future research should focus on dichotomous outcome scores.…”

Section: Conclusion and Discussionmentioning

confidence: 99%

The Design of Cluster Randomized Trials With Random Cross-Classifications

Moerbeek

Safarkhani²

2017

Journal of Educational and Behavioral Statistics

View full text Add to dashboard Cite

Data from cluster randomized trials do not always have a pure hierarchical structure. For instance, students are nested within schools that may be crossed by neighborhoods, and soldiers are nested within army units that may be crossed by mental health-care professionals. It is important that the random cross-classification is taken into account while planning a cluster randomized trial. This article presents sample size equations, such that a desired power level is achieved for the test on treatment effect. Furthermore, it also presents optimal sample sizes given a budgetary constraint, with a special focus on conditional optimal designs where one of the sample sizes is fixed beforehand. The optimal design methodology is illustrated using a postdeployment training to reduce illhealth in armed forces personnel.

show abstract

Who pays you to be green? How customers' environmental practices affect the sales benefits of suppliers' environmental practices

Chen

2019

J of Ops Management

View full text Add to dashboard Cite

This article examines whether a supplier's sales from a customer depends on the supplier's engagement in pro‐environmental practices (PEPs), and how this relationship is moderated by the customer's PEP level. This article synthesizes findings from several streams of literature including corporate environmental management, supply chain management, and contingency theory. We hypothesize that (1) a supplier gets greater sales benefits from its PEPs when the (corporate) customer's PEP level is higher, and (2) a supplier gets higher sales from a customer when the supplier's PEP level aligns with the customer's PEP level. To test the hypotheses, we use data from public U.S. manufacturing firms and their key customers between 2006 and 2016. The results show that the effect of a manufacturer's PEPs on sales is significantly higher when the customer's PEP level is higher. The results also show that a manufacturer's sales from a particular customer are higher when the manufacturer's PEP level does not exceed the customers' PEP level. This study provides empirical evidence that an investment in PEPs does not always enhance sales—the sales effect of PEPs may vary according to the customer's PEP level. To our knowledge, this is the first large‐scale study that examines the effect of environmental practices on sales.

show abstract

The impact of ignoring multiple membership data structures in multilevel models

Cited by 73 publications

References 25 publications

Estimating a Three-Level Latent Variable Regression Model With Cross-Classified Multiple Membership Data

Estimating a Three-Level Latent Variable Regression Model With Cross-Classified Multiple Membership Data

The Design of Cluster Randomized Trials With Random Cross-Classifications

Who pays you to be green? How customers' environmental practices affect the sales benefits of suppliers' environmental practices

Contact Info

Product

Resources

About