Stochastic search item selection for factor analytic models

In this paper we implement a Markov chain Monte Carlo algorithm based on the stochastic search variable selection method of George and McCulloch (1993) for identifying promising subsets of manifest variables (items) for factor analysis models. The suggested algorithm is constructed by embedding in the usual factor analysis model a normal mixture prior for the model loadings with latent indicators used to identify not only which manifest variables should be included in the model but also how each manifest variable is associated with each factor. We further extend the suggested algorithm to allow for factor selection. We also develop a detailed procedure for the specification of the prior parameters values based on the practical significance of factor loadings using ideas from the original work of George and McCulloch (1993). A straightforward Gibbs sampler is used to simulate from the joint posterior distribution of all unknown parameters and the subset of variables with the highest posterior probability is selected. The proposed method is illustrated using real and simulated data sets.     

Health Beliefs and Illness Perceptions as Related to Mammography Uptake in Randomly Selected Women in Greece

Previous research has shown that, although routine mammography screening can reduce mortality from breast cancer, the use of annual mammography screening remains well under 65%. In an effort to determine the factors that are associated with women’s mammography behaviors, this study used the health belief model and the common-sense model of self-regulation as the theoretical frameworks to explore health beliefs, illness representations, and women’s mammography practice. Data were obtained from a nationally representative sample of 408 Greek women, 40 years of age or older, with no personal history of cancer. Three dependent variables were considered: recent mammography, repeat mammography, and no mammogram during lifetime. Predictors included socio-demographic and medical variables, perceived benefits of mammography screening, perceived barriers to mammography screening, self-efficacy, as well as illness perceptions. Multivariate analyzes indicated that never having had a mammogram was more likely for women who perceived fewer benefits and more barriers to mammography screening, had more negative emotional representations of breast cancer, and had no private health insurance coverage. Factors associated with recent mammography were younger age, a good knowledge of the recommended mammography screening interval, a family history of breast cancer, and use of patient reminders for next mammogram. Adequate knowledge about the recommended mammography screening interval and higher values for breast cancer worry were associated with an increased number of repeat lifetime mammograms. Implications of the results and suggestions for future research are outlined.     

Spatial specificity of tactile enhancement during reaching

Tactile signals on a hand that serves as movement goal are enhanced during movement planning and execution. Here, we examined how spatially specific tactile enhancement is when humans reach to their own static hand. Participants discriminated two brief and simultaneously presented tactile stimuli: a comparison stimulus on the left thumb or little finger from a reference stimulus on the sternum. Tactile stimuli were presented either during right-hand reaching towards the left thumb or little finger or while holding both hands still (baseline). Consistent with our previous findings, stimuli on the left hand were perceived stronger during movement than baseline. However, tactile enhancement was not stronger when the stimuli were presented on the digit that served as reach target, thus the perception across the whole hand was uniformly enhanced. In experiment 2, we also presented stimuli on the upper arm in half of the trials to reduce the expectation of the stimulus location. Tactile stimuli on the target hand, but not on the upper arm, were generally enhanced, supporting the idea of a spatial gradient of tactile enhancement. Overall, our findings argue for low spatial specificity of tactile enhancement at movement-relevant body parts, here the target hand.     

Psychometric properties and factor structure of the Greek version of the Cardiac Anxiety Questionnaire (CAQ)

The present study assesses the psychometric properties and factor structure of the Greek version of the Cardiac Anxiety Questionnaire (CAQ). The questionnaire was administered to 598 healthy individuals from 15 different regions of Greece with a measure of socioeconomic characteristics and the Symptom Checklist-90-Revised (SCL-90-R). The sample was split into two random halves, and exploratory factor analysis indicated a three-factor solution. This solution was tested using a confirmatory factor analysis on the second half of the sample. In terms of latent dimensions, the Greek version retains the three-factor structure as proposed by the initial authors. However, adequate fit was achieved only after omitting eight items. The shorter (10-item) version was submitted to further analysis. The shorter version provided satisfactory internal reliability and evidence indicating the validity of the scale with respect to SCL-90-R subscales. The stability of the questionnaire was verified by a high test-retest reliability over a 3-mo. period (r = .86). Sex and age differences were assessed. The 10-item version appears to be a practical, brief tool for clinical use.     

The emergence, structure and development of ethnic identity during childhood: the case of Roma identity

This study investigates the emergence, development and structure of ethnic identity during childhood. Forty Roma children living in Greece aged between 2.8 and 11.9 years answered questions about their awareness/recognition of four aspects of their ethnic identity-namely place of habitation, traditional costumes, the Roma language, and early betrothal of children-their identity and their sense of stability and constancy. The study also investigates how the children feel about the abandonment of those four aspects. The evidence from the current data supports the hypothesis that awareness of ethnic identity emerges before the age of 4. Moreover, this study offers direct empirical evidence of the multidimensionality of ethnic identity. A model of three concentric rings is proposed, extending from a core containing the most highly valued aspects of ethnic identity to the outer annulus that comprises the nonpermanent and nonstable aspects of ethnic identity. The aspects in each annulus differ in terms of the development of the sense of stability and constancy and the feelings associated with loss of the aspects in question. Even the youngest participants considered the aspects in the core to be stable and constant as well as emotionally charged; and even the 11-year-olds did not consider the aspects contained in the outer, more fluid annulus as stable and constant aspects of their ethnic identity. The development of an aspect is determined by what the majority of adults in a society, at a particular time in history, consider to be most important.     

Generalized latent variable models with non‐linear effects

Until recently, item response models such as the factor analysis model for metric responses, the two‐parameter logistic model for binary responses and the multinomial model for nominal responses considered only the main effects of latent variables without allowing for interaction or polynomial latent variable effects. However, non‐linear relationships among the latent variables might be necessary in real applications. Methods for fitting models with non‐linear latent terms have been developed mainly under the structural equation modelling approach. In this paper, we consider a latent variable model framework for mixed responses (metric and categorical) that allows inclusion of both non‐linear latent and covariate effects. The model parameters are estimated using full maximum likelihood based on a hybrid integration–maximization algorithm. Finally, a method for obtaining factor scores based on multiple imputation is proposed here for the non‐linear model.     

Understanding linguistic time sequence and simultaneity: A literature review and some new data

This study investigates the difficulty order in (1) sequence comprehension versus simultaneity comprehension and production, and (2) contribution of semantics, syntax, order of mention, and cognitive (Raven's CPM) and linguistic factors (PPVT) in understanding sequence and simultaneity constructions in Greek, which display similar properties to English sequence and simultaneity constructions. In general, results with kindergartners, first-, second-, and sixth-graders, and graduate school students supported the findings of psycholinguists who claimed that Piaget has specified difficulty order of simultaneity over succession. Performance on simultaneity production tasks was lower then performance on sequence. Performance on simultaneity production tasks was better than performance on simultaneity comprehension and consistently lower than performance on sequence. No consistent order of difficulty between the different construction types of sequence and simultaneity was found. Semantics (arbitrary vs. logical sequences) seems to affect performance, as a few studies have indicated. Unlike most previous findings, an order of mention strategy influences performance when the syntax (transformationally simple vs. complex) and semantics effect is partialed out. Syntax (when not confounded with the two previous variables) does not affect performance. Cognitive (Raven's CPM) and linguistic (PPVT) factors contributed approximately equal amounts of variance to performance on sequence and simultaneity by children in kindergarten and first and second grades.Research and Evaluation Department, Ft. Worth Independent School District.     

On the Concept of Comparison in Mentally Retarded and Nonretarded Children

Two adjective pairs in comparative form, more-less and taller-shorter, were tested with moderately mentally retarded (MR) and nonretarded children (NR) matched on verbal mental age (MA) and socioeconomic status (SES) in a show and an act-out experimental condition. The data indicated that all the main effects, the group, the experimental condition, and the unmarked-marked relational terms were significant and so were the interactions of the Group × Experimental Conditions, Group × Unmarked Versus Marked Relational Terms, and Group × Experimental Conditions and Unmarked Versus Marked Relational Terms. A logistic regression analysis (LRA) showed that the relational term shorter in the act-out condition best predicted classification membership for the subjects in either group, suggesting that the MR children were less able to self-regulate behavior. The error types, however, mainly observed in the act-out experimental condition were the same from a semantic point of view in either group. The effect of the unmarked versus marked distinction cannot be considered independent of the experimental conditions applied.