Paras D. Mehta

The role of instruction in learning to read: Preventing reading failure in at-risk children.

First and 2nd graders (N = 285) receiving Title 1 services received 1 of 3 kinds of classroom reading programs: direct instruction in letter–sound correspondences practiced in decodable text (direct code); less direct instruction in systematic sound–spelling patterns embedded in connected text (embe

People are variables too: multilevel structural equations modeling.

The article uses confirmatory factor analysis (CFA) as a template to explain didactically multilevel structural equation models (ML-SEM) and to demonstrate the equivalence of general mixed-effects models and ML-SEM. An intuitively appealing graphical representation of complex ML-SEMs is introduced that succinctly describes the underlying model and its assumptions. The use of definition variables (i.e., observed variables used to fix model parameters to individual specific data values) is extended to the case of ML-SEMs for clustered data with random slopes. Empirical examples of multilevel CFA and ML-SEM with random slopes are provided along with scripts for fitting such models in SAS Proc Mixed, Mplus, and Mx. Methodological issues regarding estimation of complex ML-SEMs and the evaluation of model fit are discussed. Further potential applications of ML-SEMs are explored.

Putting the individual back into individual growth curves

Scaling of time (age) in latent growth curve (LGC) models has important implications for studies of development. When participants begin a study at different ages, sample means and covariance-based structural equation modeling (SEM) approaches produce biased estimates of the variance of the intercept and the covariance between the Intercept and Slope factors. However, individual data vector-based SEM approaches produce proper estimates of these parameters that are identical to those produced by multilevel modeling (MLM). Scaling of the time variable also raises issues regarding the interpretation of within- and between-persons effects of time that parallel those associated with centering of predictor variables in MLM. A numerical example is used to illustrate these issues, and an Mx script for fitting individual data vector-based LGC models is provided.

Coping Efficacy and Psychological Problems of Children of Divorce

Three models of the relations of coping efficacy, coping, and psychological problems of children of divorce were investigated. A structural equation model using cross-sectional data of 356 nine- to twelve-year-old children of divorce yielded results that supported coping efficacy as a mediator of the relations between both active coping and avoiding coping and psychological problems. In a prospective longitudinal model with a subsample of 162 of these children, support was found for Time 2 coping efficacy as a mediator of the relations between Time 1 active coping and Time 2 internalizing of problems. Individual growth curve models over four waves also found support for coping efficacy as a mediator of the relations between active coping and psychological problems. No support was found for alternative models of coping as a mediator of the relations between efficacy and symptoms or for coping efficacy as a moderator of the relations between coping and symptoms.

Identifying Trajectories of Adolescent Smoking: An Application of Latent Growth Mixture Modeling

The goal of the current study was to identify discrete longitudinal patterns of change in adolescent smoking using latent growth mixture modeling. Five distinct longitudinal patterns were identified. A group of early rapid escalators was characterized by early escalation (at age 13) that rapidly increased to heavy smoking. A pattern characterized by occasional puffing up until age 15, at which time smoking escalated to moderate levels was also identified (late moderate escalators). Another group included adolescents who, after age 15, began to escalate slowly in their smoking to light (0.5 cigarettes per month) levels (late slow escalators). Finally, a group of stable light smokers (those who smoked 1-2 cigarettes per month) and a group of stable puffers (those who smoked only a few puffs per month) were also identified. The stable puffer group was the largest group and represented 25% of smokers.

Literacy as a Unidimensional Multilevel Construct: Validation, Sources of Influence, and Implications in a Longitudinal Study in Grades 1 to 4

This study examined the extent to which literacy is a unitary construct, the differences between literacy and general language competence, and the relative roles of teachers and students in predicting literacy outcomes. Much of past research failed to make a distinction between variability in outcomes for individual students and variability for outcomes in the classrooms students share (i.e., the classroom level). Utilizing data from 1,342 students in 127 classrooms in Grades 1 to 4 in 17 high-poverty schools, confirmatory factor models were fit with single- and two-factor structures at both student and classroom levels. Results support a unitary literacy factor for reading and spelling, with the role of phonological awareness as an indicator of literacy declining across the grades. Writing was the least related to the literacy factor but the most impacted by teacher effects. Language competence was distinct at the student level but perfectly correlated with literacy at the classroom level. Implications for instruction and assessment of reading comprehension are discussed.

Bilingual Phonological Awareness: Multilevel Construct Validation among Spanish-Speaking Kindergarteners in Transitional Bilingual Education Classrooms.

The construct validity of English and Spanish phonological awareness (PA) tasks was examined with a sample of 812 kindergarten children from 71 transitional bilingual education program classrooms located in 3 different types of geographic regions in California and Texas. Tasks of PA, including blending nonwords, segmenting words, and phoneme elision, were measured in Spanish and in English and analyzed via multilevel confirmatory factor analysis at the task level. Results showed that the PA tasks defined a unitary construct at both the student and classroom levels in each language. English and Spanish PA factors were related to each other (.93 and .83 at the student and classroom levels, respectively) as well as to word reading, both within languages (correlations estimated between .74 and .93) and across languages (correlations estimated between .47 and .79). Although the PA constructs were statistically separable in each language, the high correlation between Spanish and English PA indicates considerable overlap in these abilities.

Squeezing Interval Change From Ordinal Panel Data: Latent Growth Curves With Ordinal Outcomes.

A didactic on latent growth curve modeling for ordinal outcomes is presented. The conceptual aspects of modeling growth with ordinal variables and the notion of threshold invariance are illustrated graphically using a hypothetical example. The ordinal growth model is described in terms of 3 nested models: (a) multivariate normality of the underlying continuous latent variables (yt) and its relationship with the observed ordinal response pattern (Yt), (b) threshold invariance over time, and (c) growth model for the continuous latent variable on a common scale. Algebraic implications of the model restrictions are derived, and practical aspects of fitting ordinal growth models are discussed with the help of an empirical example and Mx script (M. C. Neale, S. M. Boker, G. Xie, & H. H. Maes, 1999). The necessary conditions for the identification of growth models with ordinal data and the methodological implications of the model of threshold invariance are discussed.

Hierarchical Model of Vulnerabilities for Emotional Disorders

Clark and Watsons (1991) tripartite model of anxiety and depression has had a dramatic impact on our understanding of the dispositional variables underlying emotional disorders. More recently, calls have been made to examine not simply the influence of negative affectivity (NA) but also mediating factors that might better explain how NA influences anxious and depressive syndromes (e.g. Taylor, 1998; Watson, 2005). Extending preliminary projects, this study evaluated two hierarchical models of NA, mediating factors of anxiety sensitivity and intolerance of uncertainty, and specific emotional manifestations. Data provided a very good fit to a model elaborated from preliminary studies, lending further support to hierarchical models of emotional vulnerabilities. Implications for classification and diagnosis are discussed.

Identifying and predicting adolescent smokers' developmental trajectories

Very few studies have defined trajectories of smoking. In the present study, we modeled growth in adolescent smoking and empirically identified prototypical trajectories. We conceptualized escalation of smoking as a growth process and modeled rates of change and heterogeneity of these patterns using latent growth mixture modeling. The analysis identified six trajectories with low ambiguity about group membership (early rapid escalators, late rapid escalators, late moderate escalators, late slow escalators-smokers, stable puffers, and late slow escalators-puffers). A trajectory of quitters was not identified. We also examined predictors of the smoking trajectories. The predictors were assessed across the adolescent years and included variables related to smoking and other substance use, as well as a range of variables related to sociodemographic factors and mental health. Observed change in the pattern of predictors across age has implications for the mechanism of effect of these variables in relation to smoking trajectories, including predictors that differentiated among daily smokers, variables that may determine the trajectory (e.g., friends smoking), and variables that may result from the trajectory (e.g., marijuana use, less attachment to friends).