Matching Items (4)
Filtering by
- All Subjects: Multiple Imputation
- All Subjects: Psychology, Behavioral
- Creators: Aiken, Leona
- Creators: Thompson, Marilyn S
Description
Missing data are common in psychology research and can lead to bias and reduced power if not properly handled. Multiple imputation is a state-of-the-art missing data method recommended by methodologists. Multiple imputation methods can generally be divided into two broad categories: joint model (JM) imputation and fully conditional specification (FCS) imputation. JM draws missing values simultaneously for all incomplete variables using a multivariate distribution (e.g., multivariate normal). FCS, on the other hand, imputes variables one at a time, drawing missing values from a series of univariate distributions. In the single-level context, these two approaches have been shown to be equivalent with multivariate normal data. However, less is known about the similarities and differences of these two approaches with multilevel data, and the methodological literature provides no insight into the situations under which the approaches would produce identical results. This document examined five multilevel multiple imputation approaches (three JM methods and two FCS methods) that have been proposed in the literature. An analytic section shows that only two of the methods (one JM method and one FCS method) used imputation models equivalent to a two-level joint population model that contained random intercepts and different associations across levels. The other three methods employed imputation models that differed from the population model primarily in their ability to preserve distinct level-1 and level-2 covariances. I verified the analytic work with computer simulations, and the simulation results also showed that imputation models that failed to preserve level-specific covariances produced biased estimates. The studies also highlighted conditions that exacerbated the amount of bias produced (e.g., bias was greater for conditions with small cluster sizes). The analytic work and simulations lead to a number of practical recommendations for researchers.
ContributorsMistler, Stephen (Author) / Enders, Craig K. (Thesis advisor) / Aiken, Leona (Committee member) / Levy, Roy (Committee member) / West, Stephen G. (Committee member) / Arizona State University (Publisher)
Created2015
Description
This study examined the influence of childhood aggression, peer exclusion and associating with deviant peers on the development of antisocial behavior in early adolescence. To gain a stronger understanding of how these factors are associated with antisocial behavior and delinquency, multiple alternative pathways were examined based on additive, mediation and incidental models. A parallel process growth model was specified to assess whether early childhood aggression and peer exclusion (in 1st grade) and intra-individual increases in aggressive behaviors and exclusion through childhood (grades 1 to 6) are predictive of associating with deviant peers (in 7th grade) and antisocial behavior (in 8th grade). Based on a sample of 383 children (193 girls and 190 boys), results showed the strongest support for an additive effects model in which early childhood aggression, increases in aggression, increases in peer exclusion and associating with more deviant peers all predicted antisocial behavior. These findings have implications for how children's psychological adjustment is impacted by their behavioral propensities and peer relational context and the importance of examining developmental processes within and between children over time.
ContributorsEttekal, Idean (Author) / Ladd, Gary W (Thesis advisor) / Eggum, Natalie D (Committee member) / Thompson, Marilyn S (Committee member) / Arizona State University (Publisher)
Created2011
Description
Accurate data analysis and interpretation of results may be influenced by many potential factors. The factors of interest in the current work are the chosen analysis model(s), the presence of missing data, and the type(s) of data collected. If analysis models are used which a) do not accurately capture the structure of relationships in the data such as clustered/hierarchical data, b) do not allow or control for missing values present in the data, or c) do not accurately compensate for different data types such as categorical data, then the assumptions associated with the model have not been met and the results of the analysis may be inaccurate. In the presence of clustered
ested data, hierarchical linear modeling or multilevel modeling (MLM; Raudenbush & Bryk, 2002) has the ability to predict outcomes for each level of analysis and across multiple levels (accounting for relationships between levels) providing a significant advantage over single-level analyses. When multilevel data contain missingness, multilevel multiple imputation (MLMI) techniques may be used to model both the missingness and the clustered nature of the data. With categorical multilevel data with missingness, categorical MLMI must be used. Two such routines for MLMI with continuous and categorical data were explored with missing at random (MAR) data: a formal Bayesian imputation and analysis routine in JAGS (R/JAGS) and a common MLM procedure of imputation via Bayesian estimation in BLImP with frequentist analysis of the multilevel model in Mplus (BLImP/Mplus). Manipulated variables included interclass correlations, number of clusters, and the rate of missingness. Results showed that with continuous data, R/JAGS returned more accurate parameter estimates than BLImP/Mplus for almost all parameters of interest across levels of the manipulated variables. Both R/JAGS and BLImP/Mplus encountered convergence issues and returned inaccurate parameter estimates when imputing and analyzing dichotomous data. Follow-up studies showed that JAGS and BLImP returned similar imputed datasets but the choice of analysis software for MLM impacted the recovery of accurate parameter estimates. Implications of these findings and recommendations for further research will be discussed.
ested data, hierarchical linear modeling or multilevel modeling (MLM; Raudenbush & Bryk, 2002) has the ability to predict outcomes for each level of analysis and across multiple levels (accounting for relationships between levels) providing a significant advantage over single-level analyses. When multilevel data contain missingness, multilevel multiple imputation (MLMI) techniques may be used to model both the missingness and the clustered nature of the data. With categorical multilevel data with missingness, categorical MLMI must be used. Two such routines for MLMI with continuous and categorical data were explored with missing at random (MAR) data: a formal Bayesian imputation and analysis routine in JAGS (R/JAGS) and a common MLM procedure of imputation via Bayesian estimation in BLImP with frequentist analysis of the multilevel model in Mplus (BLImP/Mplus). Manipulated variables included interclass correlations, number of clusters, and the rate of missingness. Results showed that with continuous data, R/JAGS returned more accurate parameter estimates than BLImP/Mplus for almost all parameters of interest across levels of the manipulated variables. Both R/JAGS and BLImP/Mplus encountered convergence issues and returned inaccurate parameter estimates when imputing and analyzing dichotomous data. Follow-up studies showed that JAGS and BLImP returned similar imputed datasets but the choice of analysis software for MLM impacted the recovery of accurate parameter estimates. Implications of these findings and recommendations for further research will be discussed.
ContributorsKunze, Katie L (Author) / Levy, Roy (Thesis advisor) / Enders, Craig K. (Committee member) / Thompson, Marilyn S (Committee member) / Arizona State University (Publisher)
Created2016
Description
Specific cultural variables have been found to protect against the onset of alcohol, tobacco and drug use among Latino adolescents. It has been suggested that targeting similar cultural components during the treatment of drug dependence and abuse for Latino adults may also enhance the effectiveness of the intervention, although few studies have explored this hypothesis. The current study attempted to remedy this disparity by exploring the potentially protective influence of two cultural variables, ethnic pride and family traditionalism, on self-efficacy to avoid drug use following residential substance abuse treatment among 99 Hispanic and 85 non-Hispanic White males. Results of the study indicate that higher levels of ethnic pride predict greater confidence to remain abstinent from drugs following substance abuse treatment, and that this relationship is stronger among Hispanic participants than non-Hispanic White participants. Family traditionalism was not a significant predictor of drug avoidance self-efficacy for either group, suggesting that some specific cultural variables may be better targets for substance abuse treatment than others. Study limitations and future directions for research and clinical practice are discussed.
ContributorsBoyd, Stephen James (Author) / Gonzalez Castro, Felipe (Thesis advisor) / Barrera, Jr., Manuel (Committee member) / Aiken, Leona (Committee member) / Arizona State University (Publisher)
Created2011