Incorporating prior theory in covariance structure analysis: A bayesian approach

A Bayesian approach to the testing of competing covariance structures is developed. The method provides approximate posterior probablities for each model under consideration without prior specification of individual parameter distributions. The method is based on ayesian updating using cross-validated pseudo-likelihoods. Given that the observed variables are the samefor all competing models, the approximate posterior probabilities may be obtained easily from the chi square values and other known constants, using only a hand calculator. The approach is illustrated using and example which illustrates how the prior probabilities can alter the results concerning which model specification is preferred.     

Least-squares approximation of an improper correlation matrix by a proper one

An algorithm is presented for the best least-squares fitting correlation matrix approximating a given missing value or improper correlation matrix. The proposed algorithm is based upon a solution for Mosier's oblique Procrustes rotation problem offered by ten Berge and Nevels. A necessary and sufficient condition is given for a solution to yield the unique global minimum of the least-squares function. Empirical verification of the condition indicates that the occurrence of non-optimal solutions with the proposed algorithm is very unlikely. A possible drawback of the optimal solution is that it is a singular matrix of necessity. In cases where singularity is undesirable, one may impose the additional nonsingularity constraint that the smallest eigenvalue of the solution be , where is an arbitrary small positive constant. Finally, it may be desirable to weight the squared errors of estimation differentially. A generalized solution is derived which satisfies the additional nonsingularity constraint and also allows for weighting. The generalized solution can readily be obtained from the standard unweighted singular solution by transforming the observed improper correlation matrix in a suitable way.     

Using item-specific instructional information in achievement modeling

The problem of detecting instructional sensitivity (item basis) in test items is considered. An illustration is given which shows that for tests with many biased items, traditional item bias detection schemes give a very poor assessment of bias. A new method is proposed instead. This method extends item response theory (IRT) by including item-specific auxiliary measurement information related to opportunity-to-learn. Item-specific variation in measurement relations across students with varying opportunity-to-learn is allowed for.This paper was presented at the 1987 AERA meeting in Washington, DC. This research was supported by grant OERI-G-86-003 from the Office of Educational Research and Improvement, Department of Education. The author thanks Michael Hollis and Chih-fen Kao for valuable research assistance, and appreciates valuable comments made by an anonymous reviewer.     

A neuroscience model of stuttering

Using motor control systems analysis and a reductionist approach, we provide a unified model of stuttering. This model views stuttering as a momentary instability in a complex multiloop control system. The model predicts the temporal conditions under which this instability will occur. Furthermore, these temporal conditions account for the efficacy of fluency-evoking maneuvers, therapy, and the variability of speech output in stutterers.     

Relationship of testosterone to men's family functioning at mid-life: A research note

Sociobiological theory suggests that aggression coupled with altruism is highly adaptive for males in all social systems, including the family, whereas sex-role convergence theory indicates that lower levels of aggression are most adaptive for men, particularly at mid-life. The purpose of this study was to determine the role that testosterone, as a proxy for aggression, plays as an adaptive mechanism in the lives of middle-aged males. The sample used in this study consisted of 37 middle-aged (39–50 years of age) males employed in professional occupations in a midwestern metropolitan area. Testosterone levels were regressed on factors typically found to be related to male satisfaction with family life at mid-life: marital satisfaction, parent-adolescent communication, amount of emotional expressiveness, and androgynous characteristics; recent positive and negative stressors and trait anxiety were statistically controlled. Results of step-wise multiple regression analysis indicated that low levels of testosterone were significantly (R = .34, P < .01) related to enhanced marital and parental relationships and androgynous behaviors; however, emotional expressiveness was significantly related to high levels of testosterone.     

Different levels of complexity in the play-fighting by muroid rodents appear to result from different levels of intensity of attack and defense

Play-fighting in deer mice, Peromyscus maniculatus bairdii, prairie voles, Microtus ochrogaster, and montane voles, M. Montanus, was compared to that of laboratory rats, Rattus norvegicus. Play in rats appears more complex for two reasons: 1) more of the playful contacts elicit defensive behaviors, and 2) more of these defenses lead to counterattacks, and hence, role reversals between attackers and defenders. Neither high levels of defense, as shown by montane voles, nor high levels of counterattack, as shown by prairie voles, produce rat-like play-fighting. This only occurs when high rates of defense involving turning to face the attacker and counterattack are combined, as in rats. These two components are rarely combined together by deer mice, and so this species rarely exhibits rat-like play-fighting. Furthermore, playful counterattack appears to arise from playful attack, and not from an escalation of defense. These data suggest that the more complex forms of social play, such as play-fighting, have evolved, in part, via the escalation of defense in response to playful attack.     

The Children's Firesetting Interview with psychiatrically referred and nonreferred children

This study evaluated the Children's Firesetting Interview (CFI). The measure was developed to operationalize multiple domains of functioning derived from a riskfactor model of firesetting. The model poses that child, parent, and family characteristics promote firesetting and continuation of a pattern of firesetting. Major factors include curiosity about fire, involvement in and exposure to firerelated activities, and knowledge about first safety The CFI, consisting of 46 questions reflecting six a prioridimensions, was administered to 519 children (ages 6–13) recruited from nonpatient, outpatient, and inpatient samples. Internal consistency and testretest reliability of the measure were satisfactory. Criterion validity was supported by the findings that firesetters showed greater curiosity about fire, involvement in firerelated activities, exposure to models/materials, and knowledge about things that burn than did nonfiresetters. These findings did not vary as a function of the child's patient status or level of antisocial behavior. Implications for the evaluation of firesetting risk are discussed.This study was supported by grant MH-39976 from the Antisocial and Violent Behavior Branch of the National Institute of Mental Health and by a Research Scientist Development Award (MH00353) from the National Institute of Mental Health. The authors acknowledge the constructive suggestions of two anonymous reviewers.     

Behavioral correlates of developmental expressive language disorder

The association of behavior problems with preschool language disorders has been documented extensively. However, researchers have typically failed to differentiate subgroups of language-impaired children, to use observational data in documenting the behavior disorders, or to study children at the youngest ages. Using a multimodal assessment, this study examined parent-child interaction and behavior problems in a clearly defined subgroup of language-impaired children, those with developmental expressive language disorder (ELD). These children exhibit a delay in expressive language compared with receptive language and nonverbal cognitive skills. Subjects were identified and studied at the youngest age at which the disorder can be assessed. A group of ELD children, averaging 27 months of age, was contrasted with a group of normally developing children, matched for age, sex, and receptive language ability. Groups were compared on observed parent-child interactions as well as maternal responses on the Parenting Stress Index, the Eyberg Child Behavior Inventory, and a behavior-related structured interview. ELD children, when compared with normally developing children, exhibited higher levels of negative behavior and were perceived as different by their parents.Portions of these data were presented at the biennial meeting of the Society for Research in Child Development, Baltimore, April 1987, and at the annual meeting of the Society for Pediatrie Research, Anaheim, California, April 1987. This work has been supported by NIMH grant no. 1 R03 MH41603 to author Fischel, and by NICHD grant no. 1 ROI HD19245 to authors Whitehurst and Fischel. It has also been supported by grants of equipment from Commodore Business Machines, Inc., Koala Corporation, and NEC Telephones. We thank the Department of Pediatrics at the Nassau County Medical Center for the use of their facilities.     

Generativity and Erikson's “belief in the species”

Generativity is the concern for guiding and promoting the next generation through such creative behavior as parenting, teaching, mentoring, leading, and generating products and outcomes that benefit others. Erikson (1963) has argued that in order to be generative in adulthood, people must have a fundamental “belief in the species” or a faith that human progress is possible and worth working toward. The present study focuses on hope and trust concerning the self and others (Erikson's “belief in the species”). In addition, however, the study examines the relation between generativity and personality traits, in this case, dominance, leadership, self-absorption, and nurturance. A sample of 70 adults was administered (1) a series of self-report questionnaires converging on Erikson's idea of belief in the species, (2) a self-report scale assessing generativity, (3) measures of personality traits, and (4) two open-ended measures of generativity requiring subjects to describe life commitments and creative endeavors. The results provide modest support for Erikson's claim of a link between belief in the species and generativity, with significant positive correlations obtained (1) between self-report generativity and two measures of belief in species and (2) between generativity assessed through life commitments and one measure of belief in the species. In addition, the personality trait of nurturance was positively related to all three generativity measures. Problems and issues in the assessment of generativity are discussed.     

Identifying depression in children: A comparison of alternative selection criteria

This study evaluated commonly used methods of identifying depressed children and examined the extent to which the conclusions about correlates of dysfunction vary among different selection criteria. Child psychiatric patients (N=237,ages 7–12) and their parents participated and completed measures to permit identification of depressed children on the basis of three separate criteria, including selfreport scores on the Children's Depression Inventory, parent-reported scores for the same measure, and a DSM-III diagnosis of major depression. The study examined whether depressed and nondepressed children, defined separately by the different criteria, differed in a variety of depression-related symptoms, cognitive processes, and social activity. The results indicated little overlap in the persons identified as depressed on the basis of child-or parent-completed CDI scores or DSM-III diagnosis. Depressed and nondepressed children tended to differ across all domains (depression-related symptoms, cognitive processes, and social activity) for each selection criterion. However, significant differences were evident in these domains only when the selection criterion and other domains were assessed with the same informant (e.g., self-report) and hence shared a common method (rater) component. The findings underscore the potential influence that method factors may have in influencing the conclusions that are drawn regarding the correlates of childhood depression.Completion of this research was supported by a Research Scientist Development Award (MH00353) and by a grant (MH35408) from the National Institute of Mental Health. The author is grateful to Debra Bass, Antoinette Rodgers, and Todd Siegel for assistance with this project.