Random effects diagonal metric multidimensional scaling models

By assuming a distribution for the subject weights in a diagonal metric (INDSCAL) multidimensional scaling model, the subject weights become random effects. Including random effects in multidimensional scaling models offers several advantages over traditional diagonal metric models such as those fitted by the INDSCAL, ALSCAL, and other multidimensional scaling programs. Unlike traditional models, the number of parameters does not increase with the number of subjects, and, because the distribution of the subject weights is modeled, the construction of linear models of the subject weights and the testing of those models is immediate. Here we define a random effects diagonal metric multidimensional scaling model, give computational algorithms, describe our experiences with these algorithms, and provide an example illustrating the use of the model and algorithms.We would like to thank J. Douglas Carroll for early consultation of this research, and Robert I. Jennrich for commenting on an earlier draft of this paper and for help on the computational algorithms. James O. Ramsay and Forrest W. Young were instrumental in providing the example data. This work was supported in part by National Institute of Mental Health grant 1 R43 MH57559-01. We would also like to thank the anonymous referees for comments that helped to clarify our work.     

A Polychoric Instrumental Variable (PIV) Estimator for Structural Equation Models with Categorical Variables

This paper presents a new polychoric instrumental variable (PIV) estimator to use in structural equation models (SEMs) with categorical observed variables. The PIV estimator is a generalization of Bollen’s (Psychometrika 61:109–121, 1996) 2SLS/IV estimator for continuous variables to categorical endogenous variables. We derive the PIV estimator and its asymptotic standard errors for the regression coefficients in the latent variable and measurement models. We also provide an estimator of the variance and covariance parameters of the model, asymptotic standard errors for these, and test statistics of overall model fit. We examine this estimator via an empirical study and also via a small simulation study. Our results illustrate the greater robustness of the PIV estimator to structural misspecifications than the system-wide estimators that are commonly applied in SEMs. Kenneth Bollen gratefully acknowledges support from NSF SES 0617276, NIDA 1-RO1-DA13148-01, and DA013148-05A2. Albert Maydeu-Olivares was supported by the Department of Universities, Research and Information Society (DURSI) of the Catalan Government, and by grant BSO2003-08507 from the Spanish Ministry of Science and Technology. We thank Sharon Christ, John Hipp, and Shawn Bauldry for research assistance. The comments of the members of the Carolina Structural Equation Modeling (CSEM) group are greatly appreciated. An earlier version of this paper under a different title was presented by K. Bollen at the Psychometric Society Meetings, June, 2002, Chapel Hill, North Carolina.     

Family Emotional Climate and Sibling Relationship Quality: Influences on Behavioral Problems and Adaptation in Preschool-Aged Children

We examined the impact of family emotional climate and sibling relationship quality on behavioral problems and adaptation in preschool-aged children. Participants were 63 mothers with a preschool-aged child enrolled in a Southern Arizona Head Start Program. Siblings were identified as children closest in age to target child. Mothers of predominantly Mexican descent (95%) participated in home interviews during the Fall and Spring of the year children entered center-based programs. Sibling relationship quality (warmth, agonism/competition) was proposed to predict children’s adjustment (behavioral problems and adaptation) longitudinally. Results indicate that after controlling for child characteristics (temperament, child gender, birth order) and after accounting for family characteristics (family emotional expressiveness, child exposure to interparental conflict, and parental agreement on childrearing), sibling warmth made a significant and unique contribution to child adjustment as reported by mothers and teachers six months later. Findings are consistent with existing research indicating that sibling relationships impact children’s adjustment and shape young children’s lives in meaningful and marked ways. Moreover, these associations were found with an understudied sample of young children of predominantly Mexican descent in low-income families, and thus make an important contribution to knowledge in the field.     

Agreement of adolescent ratings with mother ratings and teacher ratings of ADHD symptom groups: A correlated trait-correlated method minus one analysis

This study examined the level of agreement of adolescent ratings with mother ratings, and adolescent ratings with teacher ratings of the inattention (IA) and hyperactivity/impulsivity (HI) symptom groups of ADHD. A total of 214 adolescents provided self-ratings of IA and HI, and their IA and HI were also rated by their mothers and teachers. The correlated trait-correlated method minus one model was applied, with adolescent ratings as the reference method, and the other two ratings as the non-reference methods. The findings indicated no additional variance in adolescent ratings for IA and HI that could not be accounted by mother ratings of IA and HI, respectively. In contrast, there was additional variance in adolescent ratings for IA and HI that could was not accounted by teacher ratings of IA and HI, respectively. The findings suggest that when diagnosing ADHD in adolescents, their reports of their own ADHD behaviors are not needed when mother reports of such behaviors are used.     

Agreement between Parents and Children on Ratings of Post-Concussive Symptoms Following Mild Traumatic Brain Injury

The level of parent-child agreement on post-concussive symptoms (PCS) was examined in children following mild traumatic brain injuries (TBI). As part of a larger longitudinal study, 186 children with mild TBI and 99 with orthopedic injuries (OI), from 8 to 15 years of age, were recruited prospectively. Parents and children completed the PCS Interview (PCS-I) and the Health and Behavior Inventory (HBI) at 2 weeks, 1 month, 3 months, and 12 months postinjury. Item-level correlations between child and parent ratings on both measures of PCS were significant but modest in both groups. Parent-child correlations for composite scales on the HBI and the total score on the PCS-I were significant in both groups, but somewhat higher in the OI group than in the mild TBI group. Mean symptom ratings tended to be significantly higher for children as compared to parents, especially for somatic symptoms. Parents and children display modest agreement when reporting PCS; their ratings correlate significantly, but children report higher mean levels of symptoms than parents.     

Processing biological gender and number information during Chinese pronoun resolution: ERP evidence for functional differentiation

There have been a number of behavioral and neural studies on the processing of syntactic gender and number agreement information, marked by different morpho-syntactic features during sentence comprehension. By using the event-related potential (ERP) technique, the present study investigated whether the processing of semantic gender information and the processing of notional number information can be differentiated and to what extent they might interact during Chinese pronoun resolution. The pronoun (with singular form in Experiment 1 and with plural form in Experiment 2) in a sentence matched its antecedent or mismatched it with respect to either biological gender or notional number or both. While the number mismatch elicited a P600 effect starting from 550 ms (for singular pronoun) or 400 ms (for plural pronoun) post-onset of the pronoun, the gender mismatch elicited an earlier (for singular) and larger (for both singular and plural) P600 effect. More importantly, the double mismatch produced a P600 effect identical to the effect elicited by the single gender mismatch. These results demonstrate that biological gender information and notional number information are processed differentially and have different processing priorities during Chinese pronoun resolution.     

A Comparison of Imputation Strategies for Ordinal Missing Data on Likert Scale Variables

This article compares a variety of imputation strategies for ordinal missing data on Likert scale variables (number of categories = 2, 3, 5, or 7) in recovering reliability coefficients, mean scale scores, and regression coefficients of predicting one scale score from another. The examined strategies include imputing using normal data models with naïve rounding/without rounding, using latent variable models, and using categorical data models such as discriminant analysis and binary logistic regression (for dichotomous data only), multinomial and proportional odds logistic regression (for polytomous data only). The result suggests that both the normal model approach without rounding and the latent variable model approach perform well for either dichotomous or polytomous data regardless of sample size, missing data proportion, and asymmetry of item distributions. The discriminant analysis approach also performs well for dichotomous data. Naïvely rounding normal imputations or using logistic regression models to impute ordinal data are not recommended as they can potentially lead to substantial bias in all or some of the parameters.