Conference on Advanced Psychometric Methods in Cognitive Aging Research

点击翻译按钮获取中文摘要

PROJECT SUMMARY ABSTRACT The rapid increase in the older segment of the population has stimulated intense scientific interest in normal and pathological cognitive functioning in older adults. The increasing prevalence with advancing age of debilitating diseases like Alzheimer's disease has led to a need for clinical and research methods for measurement of effects of these diseases, and cognitive tests play a major role in detecting, diagnosing and monitoring disease status and progression. Increasing demographic diversity creates special challenges for accurate measurement of cognition. Most cognitive tests that are in clinical and research use were developed using psychometric methods from the first half of the 20th century. There have been substantial advances in measurement theory and methodology, notably item response theory (IRT) and associated latent variable modeling methods, that could have an important impact on the measurement of cognition. There have been parallel advances in statistical methodology for modeling longitudinal cognitive trajectories and identifying variables that positively and negatively impact these important outcomes. This conference series is designed to promote the application of modern psychometric and statistical methods in research on cognitive aging. Specific goals are: 1) to expose developing and established researchers in cognitive aging to modern psychometric and statistical modeling techniques, 2) to expose experts in psychometric theory and statistics to the practical and theoretical concerns of cognitive aging research, and 3) encourage collaborations of researchers, psychometricians, and statisticians during these conferences. We have conducted nine successful annual conferences since initial funding of this grant in 2008 and have a 10th planned for September, 2017. This series has generated 59 collaborative publications to date. This experience has helped to shape our plans for the next generation of conferences. The format of the conferences will include didactic presentations by experts in cognitive aging and applied psychometric theory, demonstrations of psychometric and statistical analytic methods, and most importantly, hands-on experience using real data. This content and format is not only appropriate for encouraging education and collaboration of seasoned researchers but has also been an extremely effective learning environment for graduate students, postdoctoral fellows, and junior faculty. Conference themes for the proposed five-year renewal period are: 1) Cognitive aging in demographically diverse populations, 2), Harmonization, linking and equating cognitive measurements, 3) Missing data, selection, and quantitative bias analysis, 4) Cognitive reserve/resilience, and 5) Hierarchical Bayesian item response theory modeling to address challenges in longitudinal data. There will be a heavy emphasis on workgroups organized around scientific analyses of real data, and we plan to disseminate the information resulting from these workshops in both traditional academic outlets (e.g. submission of manuscripts for peer reviewed scientific journals) and through symposia at annual scientific conferences.