			Home About us Contact

Very Large Sample Sizes (very + large_sample_size)

Distribution by Scientific Domains

Medical Sciences	50%

Selected Abstracts

Formulation of the DETECT Population Parameter and Evaluation of DETECT Estimator Bias

JOURNAL OF EDUCATIONAL MEASUREMENT, Issue 3 2006
Louis A. Roussos
The development of the DETECT procedure marked an important advancement in nonparametric dimensionality analysis. DETECT is the first nonparametric technique to estimate the number of dimensions in a data set, estimate an effect size for multidimensionality, and identify which dimension is predominantly measured by each item. The efficacy of DETECT critically depends on accurate, minimally biased estimation of the expected conditional covariances of all the item pairs. However, the amount of bias in the DETECT estimator has been studied only in a few simulated unidimensional data sets. This is because the value of the DETECT population parameter is known to be zero for this case and has been unknown for cases when multidimensionality is present. In this article, integral formulas for the DETECT population parameter are derived for the most commonly used parametric multidimensional item response theory model, the Reckase and McKinley model. These formulas are then used to evaluate the bias in DETECT by positing a multidimensional model, simulating data from the model using a very large sample size (to eliminate random error), calculating the large-sample DETECT statistic, and finally calculating the DETECT population parameter to compare with the large-sample statistic. A wide variety of two- and three-dimensional models, including both simple structure and approximate simple structure, were investigated. The results indicated that DETECT does exhibit statistical bias in the large-sample estimation of the item-pair conditional covariances; but, for the simulated tests that had 20 or more items, the bias was small enough to result in the large-sample DETECT almost always correctly partitioning the items and the DETECT effect size estimator exhibiting negligible bias. [source]

Analysis of Genetically Complex Epilepsies

EPILEPSIA, Issue 2005
Ruth Ottman
Summary:, During the last decade, great progress has been made in the discovery of genes that influence risk for epilepsy. However, these gene discoveries have been in epilepsies with Mendelian modes of inheritance, which comprise only a tiny fraction of all epilepsy. Most people with epilepsy have no affected relatives, suggesting that the great majority of all epilepsies are genetically complex: multiple genes contribute to their etiology, none of which has a major effect on disease risk. Gene discovery in the genetically complex epilepsies is a formidable task. It is unclear which epilepsy phenotypes are most advantageous to study, and chromosomal localization and mutation detection are much more difficult than in Mendelian epilepsies. Association studies are very promising for the identification of complex epilepsy genes, but we are still in the earliest stages of their application in the epilepsies. Future studies should employ very large sample sizes to ensure adequate statistical power, clinical phenotyping methods of the highest quality, designs and analytic techniques that control for population stratification, and state-of-the-art molecular methods. Collaborative studies are essential to achieve these goals. [source]

Study Design in Osteoporosis: A European Perspective,

JOURNAL OF BONE AND MINERAL RESEARCH, Issue 6 2003
Ja Kanis MD
The advent of effective agents for the treatment of osteoporosis has led to the view that placebo-controlled trials to test new agents for efficacy are no longer appropriate. Rather, studies of superiority, equivalence, or non-inferiority have been recommended. Such studies require very large sample sizes, and the burden of osteoporotic fracture in a trial setting is substantially increased. Studies of equivalence cannot be unambiguously interpreted because the variance in effect of active comparator agents is too large in osteoporosis. If fracture studies are required by regulatory agencies, there is still a requirement for placebo-controlled studies, although perhaps of shorter duration than demanded at present. [source]

Few beetle species can be detected with 95% confidence using pitfall traps

AUSTRAL ECOLOGY, Issue 1 2010
DON A. DRISCOLL
Abstract False absences in wildlife surveys make it difficult to identify metapopulation processes, increase uncertainty of management decisions and bias parameter estimates in habitat models. Despite these risks, the number of species that can be detected with a certain probability in a community survey has rarely been examined. I sampled beetles over 5 months using pitfall trap grids at three rainforest locations in Tasmania, Australia. I compared detection probability for dispersed and clustered sampling schemes using a zero-inflated binomial model and a simpler occurrence method to calculate the probability of detection. After excluding extremely rare species, I analysed 12 of 121 species. Only three to six species could be detected with 95% probability using a sampling effort that is frequently applied in ecological studies. A majority of common species had a mid summer peak in detection probability meaning that survey effort could be reduced from 5 to 2 months with only a small reduction in data quality. Most species occurred at only a proportion of sample points within locations. Despite the implied spatial structuring, three small grids within a location detected 10 of 12 species as effectively as large, dispersed grids. This study warns that as little as 5% of the beetle fauna may have a 95% probability of detection using the frequently applied pitfall trap method, highlighting a substantial limitation in our ability to accurately map the distributions of ground invertebrates. Whether very large sample sizes can overcome this limitation remains to be examined. [source]