Random coefficient models for binary longitudinal responses with attrition

We extend the approach introduced by Aitkin and Alfò (1998, Statistics and Computing, 4, pp. 289–307) to the general framework of random coefficient models and propose a class of conditional models to deal with binary longitudinal responses, including unknown sources of heterogeneity in the regression parameters as well as serial dependence of Markovian form.Furthermore, we discuss the extension of the proposed approach to the analysis of informative drop-outs, which represent a central problem in longitudinal studies, and define, as suggested by Follmann and Wu (1995, Biometrics, 51, pp. 151–168), a conditional specification of the full shared parameter model for the primary response and the missingness indicator. The model is applied to a dataset from a methadone maintenance treatment programme held in Sydney in 1986 and previously analysed by Chan et al. (1998, Australian & New Zealand Journal of Statistics, 40, pp. 1–10).All of the proposed models are estimated by means of an EM algorithm for nonparametric maximum likelihood, without assuming any specific parametric distribution for the random coefficients and for the drop-out process.A small scale simulation work is described to explore the behaviour of the extended approach in a number of different situations where informative drop-outs are present.     

Multiple Measures of Fixation on Social Content in Infancy: Evidence for a Single Social Cognitive Construct?

The preference of infants to fixate on social information in a stimulus is well known. We examine how this preference manifests across a series of free‐viewing tasks using different stimulus types. Participants were thirty typically developing infants. We measured eye movements when viewing isolated faces, faces alongside objects in a grid, and faces naturally presented in photographed scenes. In each task, infants fixated social content for longer than nonsocial content. Social preference scores representing distribution of fixation to social versus general image content were highly correlated and thus combined into a single composite measure, which was independent of demographic and behavioral measures. We infer that multiple eye‐tracking tasks can be used to generate a composite measure of social preference in infancy. This approach may prove useful in the early characterization of developmental disabilities.     

Kinship Caregiver Perception of a State-Supervised Kinship Navigator Program

This study evaluated a state-administered Kinship Navigator (KN) program providing information and support to grandparents and other relatives caring for children who are not in child welfare custody. Caregivers’ perceptions of types of services, impact, satisfaction, and areas for improvement were examined. Caregivers (N = 92) reported feeling supported, gaining knowledge about services, and feeling highly satisfied with the program. Caregivers expressed needs for additional financial assistance and more time with their KN. Findings suggest that the KN program may meet key service needs of informal kinship caregivers, but additional staff time and financial resource advocacy would improve the program.     

RAPID CONVERGENCE TECHNIQUES FOR MARKOV DECISION PROCESSES*

When a person is working with large scale Markov Decision Processes, he normally uses the policy iteration approach developed by Howard [1] and modified by White [3]. White's modification makes use of the method of successive approximations. Computational experience has shown that for many processes, the rate of convergence of the successive approximation is very slow. In this paper, techniques for speeding convergence are discussed. Numerical examples and computational experience which show the relative merits of the various approaches are presented.     

Estimation of Missing Cells in Randomized Block and Latin Square Designs

Classical missing cell formulas for the randomized block design and Latin square design are derived using the modified cell means model. The role of the assumptions of no interaction for these designs is made clear in the development. Examples are given for the randomized block and Latin square.     

The Effect of Migration on a Similarity Index

Fewster and Buckland (2001 Fewster , R. M. , Buckland , S. T. ( 2001 ). Similarity indices for spatial ecological data . Biometrics 57 : 495 – 501 . [Google Scholar]) defined a similarity index between two communities by allowing changes between sites to reduce the influence of local discrepancies. The similarity index of Fewster and Buckland is calculated to attain the maximum similarity between two communities in the presence of migration. Instead of maximizing similarity, we propose random migration to measure the similarity of two communities with two types of stochastic migration. The similarity values based on the proposed methods can be treated as the expected value of similarity under migration. We use computer simulation and empirical examples to demonstrate our approach.