Decomposing the total variation in a nested random effects model of neighborhood, household, and individual components when the dependent variable is dichotomous: implications for adolescent marijuana use

Wright, Douglas A.; Bobashev, Georgiy; Novak, Scott P.

doi:10.1016/j.drugalcdep.2004.11.002

Cited by 12 publications

(6 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Including time in this model should not alter the variance structure of the outcome because time is not clustered within communities. Evidence of between-person and betweencommunity variation at this step justifies our addressing individual-level and community-level predictors that may "explain" variation in the intercept (Bryk & Raudenbush, 1992;Diez Roux, 2004;Wright, Bobashev, & Novak, 2005). Cross-level interactions were tested during model building to evaluate whether the effects of community-level characteristics differ by demographic characteristics of residents; and the interaction between urbanization and time was tested to evaluate whether the impact of urbanization on the outcome differed by time period.…”

Section: Discussionmentioning

confidence: 99%

China's transition: The effect of rapid urbanization on adult occupational physical activity

Monda

Gordon‐Larsen

Stevens

et al. 2007

Social Science & Medicine

213

185

View full text Add to dashboard Cite

China has recently undergone rapid social and economic change. Increases in urbanization have led to equally rapid shifts toward more sedentary occupations through the acquisition of new technology and transitions away from a mostly agricultural economy. Our purpose was to utilize a detailed measure of urbanicity comprising 10 dimensions of urban services and infrastructure to examine its effects on the occupational physical activity patterns of Chinese adults. Longitudinal data were from individuals aged 18-55 from the years 1991-1997 of the China Health and Nutrition Survey (N=4376 men and 4384 women). Logistic multilevel regression analyses indicated that men had 68% greater odds, and women had 51% greater odds, of light versus heavy occupational activity given the mean change in urbanization over the 6-year period. Further, simulations showed that light occupational activity increased linearly with increasing urbanization. After controlling for individual-level predictors, community-level urbanization explained 54% and 40% of the variance in occupational activity for men and women, respectively. This study provides empirical evidence of the reduction in intensity of occupational activity with modernization. It is likely that urbanization will continue unabated in China and this is liable to lead to further transitions in the labor market resulting in additional reductions in work-related activity. Because occupational activity remains the major source of energy expenditure for adults, unless alternative forms are widely adopted, the Chinese population is at risk of dramatic increases in the numbers of overweight and obese individuals.

show abstract

Section: Discussionmentioning

confidence: 99%

China's transition: The effect of rapid urbanization on adult occupational physical activity

Monda

Gordon‐Larsen

Stevens

et al. 2007

Social Science & Medicine

213

185

View full text Add to dashboard Cite

show abstract

“…It is possible then to determine the proportion of variation in marital disruption that occurs across individuals versus that which occurs within individuals. Although there are alternatives (Wright, Bobashev, & Novak, 2004), the simplest strategy is to assume a latent‐variable approach (assuming that the binary outcome, marital dissolution, is simply the realization of a latent, unobserved propensity). In that case the proportion of variance that occurs within individuals is ), where π 2 /3 is the variance of the standard logistic distribution ( π = 3.29) and is the variance estimated when only an intercept term is fitted ( estimated with no covariates).…”

Section: Random Effectsmentioning

confidence: 99%

Modeling Repeatable Events Using Discrete-Time Data: Predicting Marital Dissolution

Teachman

2011

Journal of Marriage and Family

View full text Add to dashboard Cite

I join two methodologies by illustrating the application of multilevel modeling principles to hazard‐rate models with an emphasis on procedures for discrete‐time data that contain repeatable events. I demonstrate this application using data taken from the 1995 National Survey of Family Growth (NSFG) to ascertain the relationship between multiple covariates and risk of subsequent marital dissolution. I consider both fixed‐ and random‐effects versions of the multilevel model, as well as a Generalized Estimating Equation alternative to estimating random effects. I compare results obtained from the various estimators, noting why differences occur, and recommend when to choose the various alternatives. I also provide a set of SAS and STATA programs that can be used to analyze the NSFG data.

show abstract

“…Marginalized logistic regression provides a modicum of control, but as in random effects logistic regression, marginal correlations are not easily determined from variance components. Another difficulty in using random effects logistic regression to estimate correlation parameters owes to the fact that the variance components are on the log-odds scale whereas the error variance is on the probability scale; a simulation method has been suggested to circumvent this problem [30,31].…”

Section: Justification: Population-averaged Model Approach For Clustementioning

confidence: 99%

Finite sample adjustments in estimating equations and covariance estimators for intracluster correlations

Preisser

Lü

Qaqish

2008

Statistics in Medicine

102

View full text Add to dashboard Cite

Bias-corrected covariance estimators are introduced in the context of an estimating equations approach for intracluster correlations among binary outcomes. Simulation study results show that the bias-corrected covariance estimators perform better than uncorrected sandwich estimators in terms of bias and coverage probabilities. Additionally, introduction of a matrix-based bias-correction into the estimating equations considerably improves point and interval estimation for the intracluster correlations. The methods are illustrated using data from a nested cross-sectional cluster trial on reducing underage drinking.

show abstract

Decomposing the total variation in a nested random effects model of neighborhood, household, and individual components when the dependent variable is dichotomous: implications for adolescent marijuana use

Cited by 12 publications

References 30 publications

China's transition: The effect of rapid urbanization on adult occupational physical activity

China's transition: The effect of rapid urbanization on adult occupational physical activity

Modeling Repeatable Events Using Discrete-Time Data: Predicting Marital Dissolution

Finite sample adjustments in estimating equations and covariance estimators for intracluster correlations

Contact Info

Product

Resources

About