Predictors of word-level literacy amongst Grade 3 children in five diverse languages

Groups of Grade 3 children were tested on measures of word-level literacy and undertook tasks that required the ability to associate sounds with letter sequences and that involved visual, auditory and phonological-processing skills. These groups came from different language backgrounds in which the language of instruction was Arabic, Chinese, English, Hungarian or Portuguese. Similar measures were used across the groups, with tests being adapted to be appropriate for the language of the children. Findings indicated that measures of decoding and phonological-processing skills were good predictors of word reading and spelling among Arabic- and English-speaking children, but were less able to predict variability in these same early literacy skills among Chinese- and Hungarian-speaking children, and were better at predicting variability in Portuguese word reading than spelling. Results were discussed with reference to the relative transparency of the script and issues of dyslexia assessment across languages. Overall, the findings argue for the need to take account of features of the orthography used to represent a language when developing assessment procedures for a particular language and that assessment of word-level literacy skills and a phonological perspective of dyslexia may not be universally applicable across all language contexts. Copyright (C) 2008 John Wiley & Sons, Ltd.

Beyond the middle: A report about literacy and numeracy development of target group students in the middle years of schooling: Volume 1

The report was commissioned by the Department of Education, Science and Training to investigate the perceived efficacy of middle years programmes in all States and Territories in improving the quality of teaching, learning and student outcomes, especially in literacy and numeracy and for student members of particular target groups. These target groups included students from lower socio-economic communities, Aboriginal and Torres Strait Islander communities, students with a language background other than English, rural and remote students, and students struggling with the transition from middle/upper primary to the junior secondary years. The project involved large scale national and international literature reviews on Australian and international middle years approaches as well as an analysis of key literacy and numeracy teaching and learning strategies being used. In the report, there is emergent evidence of the relative efficacy of a combination of explicit state policy, dedicated funding and curriculum and professional development frameworks that are focused on the improvement of classroom pedagogy in the middle years. The programs that evidenced the greatest current and potential value for target group students tended to have developed in state policy environments that encouraged a structural rather than adjunct approach to middle years innovations. The authors conclude that in order to translate the gains made into sustainable improvement of educational results in literacy and numeracy for target groups, there is a need for a second generation of middle years theorising, research, development and practice.

Beyond the middle: A report about literacy and numeracy development of target group students in the middle years of schooling: Volume 2 (Appendices)

This research project was commissioned by the Commonwealth Department of Education, Science and Training to investigate the perceived efficacy of middle years programs in all States and Territories in improving the quality of teaching, learning and student outcomes - especially in literacy and numeracy and for student members of particular target groups. The latter groups included students from lower socio-economic communities, Aboriginal and Torres Strait Islander (Indigenous) communities, students with a Language Background Other than English (hereafter LBOTE), rural and remote students, and students struggling with the transition from middle/upper primary to the junior secondary years.

An evaluation of the changes in malocclusion index scores over a 25-year period

Background: The measured values of specific traits of occlusion may be subject to significant change due to growth and maturation of the dentofacial structures. Some traits may show improvement while others may show deterioration. Rarely is there an opportunity to examine a sample of occlusions 25 years after the acquisition of the original set of records. This study examines the changes in traits of occlusion in a sample of 46 subjects who were originally examined between 1971-1973 and for whom records were again obtained in 1998. Methods: The 46 patients were a sub-group of a previously selected randomised school-based sample and study models obtained in 1971-1973 were still available. New models for each patient were obtained in 1998. Of the 46 subjects, only eight had received orthodontic treatment. Results: Assessments of the changes in specific traits were made using the methods proposed in the Harry L Draker, California Modification (HLD Cal Mod) index. This simple index was chosen because the main component traits were well defined and, when analysed separately, reflected changes with time. The total index score gave a broad indication of the global changes in the individual's occlusion. The five basic traits of the HLD index include overjet, overbite, openbite, mandibular protrusion and labio-lingual spread. Three additional traits (ectopic eruption, anterior crowding and posterior crossbite) are used in the HLD Cal Mod index. These traits provided a useful reflection of occlusal changes with time. Measurements were made with reference to specifications and the details outlined in the HLD Cal Mod protocol. The results revealed an increase in total index scores over time with a significant increase in lower labio-lingual spread associated with an increased score in anterior crowding. Overjet and overbite, however, displayed a significant decrease with time. Conclusions: These findings are in keeping with previous studies and highlight the importance of time as a significant issue in the assessment of occlusion.

Clinical assessment of HIV-associated lipodystrophy syndrome: bioelectrical impedance analysis, anthropometry and clinical scores

Background Diagnosis of the HIV-associated lipodystrophy syndrome is based on clinical assessment, in lack of a consensus about case definition and reference methods. Three bedside methods were compared in their diagnostic value for lipodystrophy. Patients and Methods. Consecutive HIV-infected outpatients (n = 278) were investigated, 128 of which also had data from 1997 available. Segmental bioelectrical impedance analysis (BIA) and waist, hip and thigh circumferences were performed. Changes in seven body regions were rated by physicians and patients using linear analogue scale assessment (LASA). Diagnostic cut-off values were searched by receiver operator characteristics. Results. Lipodystrophy was diagnosed in 85 patients (31%). BIA demonstrated higher fat-free mass in patients with lipodystrophy but not after controlling for body mass index and sex. Segmental BIA was not superior to whole body BIA in detecting lipodystrophy. Fat-free mass increased from 1997 to 1999 independent from lipodystrophy. Waist-hip and waist-thigh ratios were higher in patients with lipodystrophy. BIA, anthropometry and LASA did not provide sufficient diagnostic cut-off values for lipodystrophy. Agreement between methods, and between patient and physician rating, was poor. Conclusion: These methods do not fulfil the urgent need for quantitative diagnostic tools for lipodystrophy. BIA estimates of fat free mass may be biased by lipodystrophy, indicating a need for re-calibration in HIV infected populations. (C) 2001 Harcourt Publishers Ltd.

Estimating and evaluating the statistics of gapped local-alignment scores

We present a novel maximum-likelihood-based algorithm for estimating the distribution of alignment scores from the scores of unrelated sequences in a database search. Using a new method for measuring the accuracy of p-values, we show that our maximum-likelihood-based algorithm is more accurate than existing regression-based and lookup table methods. We explore a more sophisticated way of modeling and estimating the score distributions (using a two-component mixture model and expectation maximization), but conclude that this does not improve significantly over simply ignoring scores with small E-values during estimation. Finally, we measure the classification accuracy of p-values estimated in different ways and observe that inaccurate p-values can, somewhat paradoxically, lead to higher classification accuracy. We explain this paradox and argue that statistical accuracy, not classification accuracy, should be the primary criterion in comparisons of similarity search methods that return p-values that adjust for target sequence length.