Literature DB >> 29885555

Does an uneven sample size distribution across settings matter in cross-classified multilevel modeling? Results of a simulation study.

Carly E Milliren1, Clare R Evans2, Tracy K Richmond3, Erin C Dunn4.   

Abstract

BACKGROUND: Recent advances in multilevel modeling allow for modeling non-hierarchical levels (e.g., youth in non-nested schools and neighborhoods) using cross-classified multilevel models (CCMM). Current practice is to cluster samples from one context (e.g., schools) and utilize the observations however they are distributed from the second context (e.g., neighborhoods). However, it is unknown whether an uneven distribution of sample size across these contexts leads to incorrect estimates of random effects in CCMMs.
METHODS: Using the school and neighborhood data structure in Add Health, we examined the effect of neighborhood sample size imbalance on the estimation of variance parameters in models predicting BMI. We differentially assigned students from a given school to neighborhoods within that school's catchment area using three scenarios of (im)balance. 1000 random datasets were simulated for each of five combinations of school- and neighborhood-level variance and imbalance scenarios, for a total of 15,000 simulated data sets. For each simulation, we calculated 95% CIs for the variance parameters to determine whether the true simulated variance fell within the interval.
RESULTS: Across all simulations, the "true" school and neighborhood variance parameters were estimated 93-96% of the time. Only 5% of models failed to capture neighborhood variance; 6% failed to capture school variance.
CONCLUSIONS: These results suggest that there is no systematic bias in the ability of CCMM to capture the true variance parameters regardless of the distribution of students across neighborhoods. Ongoing efforts to use CCMM are warranted and can proceed without concern for the sample imbalance across contexts.
Copyright © 2018 Elsevier Ltd. All rights reserved.

Entities:  

Keywords:  Contextual effects; Cross-classified multilevel modeling; Neighborhoods; Schools; Simulation

Mesh:

Year:  2018        PMID: 29885555      PMCID: PMC6171360          DOI: 10.1016/j.healthplace.2018.05.009

Source DB:  PubMed          Journal:  Health Place        ISSN: 1353-8292            Impact factor:   4.078


  10 in total

Review 1.  Multilevel analyses of neighbourhood socioeconomic context and health outcomes: a critical review.

Authors:  K E Pickett; M Pearl
Journal:  J Epidemiol Community Health       Date:  2001-02       Impact factor: 3.710

Review 2.  The neighborhoods they live in: the effects of neighborhood residence on child and adolescent outcomes.

Authors:  T Leventhal; J Brooks-Gunn
Journal:  Psychol Bull       Date:  2000-03       Impact factor: 17.737

Review 3.  Multilevel analysis in public health research.

Authors:  A V Diez-Roux
Journal:  Annu Rev Public Health       Date:  2000       Impact factor: 21.981

Review 4.  Multilevel models: applications to health data.

Authors:  N Rice; A Leyland
Journal:  J Health Serv Res Policy       Date:  1996-07

5.  Age differences in the association of childhood obesity with area-level and school-level deprivation: cross-classified multilevel analysis of cross-sectional data.

Authors:  N Townsend; H Rutter; C Foster
Journal:  Int J Obes (Lond)       Date:  2011-10-18       Impact factor: 5.095

Review 6.  Is there a "school effect" on pupil outcomes? A review of multilevel studies.

Authors:  E Sellström; S Bremberg
Journal:  J Epidemiol Community Health       Date:  2006-02       Impact factor: 3.710

7.  School-level contextual influences on smoking and drinking among Asian and Pacific Islander adolescents.

Authors:  Jinsook Kim; William J McCarthy
Journal:  Drug Alcohol Depend       Date:  2006-01-18       Impact factor: 4.492

8.  Social capital and adolescent smoking in schools and communities: a cross-classified multilevel analysis.

Authors:  Bart De Clercq; Timo-Kolja Pfoertner; Frank J Elgar; Anne Hublet; Lea Maes
Journal:  Soc Sci Med       Date:  2014-08-15       Impact factor: 4.634

9.  Multiple contexts and adolescent body mass index: Schools, neighborhoods, and social networks.

Authors:  Clare R Evans; Jukka-Pekka Onnela; David R Williams; S V Subramanian
Journal:  Soc Sci Med       Date:  2016-06-03       Impact factor: 4.634

Review 10.  Prevalence of obesity in the United States.

Authors:  M L Baskin; J Ard; F Franklin; D B Allison
Journal:  Obes Rev       Date:  2005-02       Impact factor: 9.213

  10 in total
  3 in total

1.  Mammals adjust diel activity across gradients of urbanization.

Authors:  Travis Gallo; Mason Fidino; Brian Gerber; Adam A Ahlers; Julia L Angstmann; Max Amaya; Amy L Concilio; David Drake; Danielle Gay; Elizabeth W Lehrer; Maureen H Murray; Travis J Ryan; Colleen Cassady St Clair; Carmen M Salsbury; Heather A Sander; Theodore Stankowich; Jaque Williamson; J Amy Belaire; Kelly Simon; Seth B Magle
Journal:  Elife       Date:  2022-03-31       Impact factor: 8.713

2.  Disentangling individual, school, and neighborhood effects on screen time among adolescents and young adults in the United States.

Authors:  Hoda S Abdel Magid; Carly E Milliren; Kelley Pettee Gabriel; Jason M Nagata
Journal:  Prev Med       Date:  2020-12-07       Impact factor: 4.018

3.  Adolescent individual, school, and neighborhood influences on young adult hypertension risk.

Authors:  Hoda S Abdel Magid; Carly E Milliren; Kathryn Rice; Nina Molanphy; Kennedy Ruiz; Holly C Gooding; Tracy K Richmond; Michelle C Odden; Jason M Nagata
Journal:  PLoS One       Date:  2022-04-28       Impact factor: 3.752

  3 in total

北京卡尤迪生物科技股份有限公司 © 2022-2023.