Matching Items (3)
Filtering by
- All Subjects: Adolescence
- All Subjects: Multiple Imputation
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
Given the major investment young people make in earning and maintaining a peer reputation, our goal in this study was to explore the association between dimensions of negative and positive peer reputation in middle school and adjustment several years later, by the end of high school, among upper middle class youth. Prior research has shown negative reputations such as aggressive-disruptive and sensitive-isolated to be associated with maladjustment later in life, whereas reputations like popular and prosocial-leader have been related to positive future outcomes. However, there are contrary findings that reveal a more complex relationship between peer reputation and adjustment, showing certain “negative” reputations to be tied with better outcomes in some domains and the converse in others. Using a sample of middle school students, a confirmatory factor analysis (CFA) was performed to test a four-factor model of the Revised Class Play, a peer report measure on peer reputations. CFA findings supported the four-factor model with the following reputations: popular, prosocial, aggressive, and isolated. Structural equation models were used to predict 12th grade adjustment outcomes (academic achievement, psychopathology, substance use) from middle school peer reputation. Prosocial reputation in middle school was connected to higher academic achievement and fewer externalizing symptoms in 12th grade. Both prosocial and isolated peer reputation were negatively associated with alcohol, cigarette, and marijuana use, whereas a popular reputation was related to higher levels of alcohol use. Middle school reputation did not predict internalizing symptoms in 12th grade. Findings are discussed in terms of adaptive and maladaptive adjustment outcomes associated with each peer reputation and implications for future research.
ContributorsCurlee, Alexandria (Author) / Luthar, Suniya (Thesis advisor) / Aiken, Leona (Committee member) / Infurna, Frank (Committee member) / Arizona State University (Publisher)
Created2016
Description
Sexual risk taking is prevalent in adolescence, particularly among Latino teens, and can have serious consequences in the form of contraction of STIs, HIV, and increased risk of unintended pregnancy. Family contexts characterized by conflict and lack of support are antecedents of adolescent sexual risk taking, but evidence elucidating the mechanisms underlying this association is lacking. The current study sought to test two potential pathways to sexual risk taking within the framework of social developmental theory, among a sample of 189 Mexican origin adolescents and their caregivers interviewed in the 7th, 8th, and 12th grades. Structural equation modeling was utilized to examine pathways from 7th grade family risk to age of sexual initiation, number of lifetime sexual partners, and condom nonuse reported in the 12th grade. Deviant peer affiliations and academic engagement at 8th grade were tested as mediators of this relationship for boys and girls. Results confirm the importance of the family context, with family risk exerting direct effects on the number of lifetime sexual partners for both genders, and on age of sexual initiation for females only. Deviant peer affiliations serve as a mediator of family risk for males, but not females. When included in a model alongside deviant peers, academic engagement does not play the hypothesized mediating role between family risk and any of the sexual risk outcomes. Future research ought to consider additional mediators that better account for the relation between family risk and sexual risk taking among females.
ContributorsJensen, Michaeline R (Author) / Gonzales, Nancy A. (Thesis advisor) / Lopez, Vera (Committee member) / Millsap, Roger (Committee member) / Arizona State University (Publisher)
Created2012