Home About us Contact
|
Home About us Contact | ||
|
|
||
General Linear Mixed Model (general + linear_mixed_model)
Selected AbstractsUse of resampling to select among alternative error structure specifications for GLMM analyses of repeated measurements,INTERNATIONAL JOURNAL OF METHODS IN PSYCHIATRIC RESEARCH, Issue 1 2004Scott Tonidandel Abstract Autocorrelated error and missing data due to dropouts have fostered interest in the flexible general linear mixed model (GLMM) procedures for analysis of data from controlled clinical trials. The user of these adaptable statistical tools must, however, choose among alternative structural models to represent the correlated repeated measurements. The fit of the error structure model specification is important for validity of tests for differences in patterns of treatment effects across time, particularly when maximum likelihood procedures are relied upon. Results can be affected significantly by the error specification that is selected, so a principled basis for selecting the specification is important. As no theoretical grounds are usually available to guide this decision, empirical criteria have been developed that focus on model fit. The current report proposes alternative empirical criteria that focus on bootstrap estimates of actual type I error and power of tests for treatment effects. Results for model selection before and after the blind is broken are compared. Goodness-of-fit statistics also compare favourably for models fitted to the blinded or unblinded data, although the correspondence to actual type I error and power depends on the particular fit statistic that is considered. Copyright © 2004 Whurr Publishers Ltd. [source] Biogeographical determinants of lichen species diversity on islets in the West-Estonian ArchipelagoJOURNAL OF VEGETATION SCIENCE, Issue 2 2006Inga Jüriado Abstract Questions: Do islet area, number of biotopes and distance from the mainland shape lichen species richness on islets? Are there any species- or substrate group-specific trends associated with these factors? Location: Islets of the west Estonian Archipelago, Estonia. Methods: A species list was compiled for each of the 32 islets and the relative abundance of each species was estimated. The lichens were divided into seven groups according to their substrate preferences. Generalized linear model (GLIM) analysis was applied to test the effect of the islet traits on the number of lichen species on the islets and in the substrate groups. The probability of presence/absence and abundance of the most frequent species according to the islet traits were tested with GLIM and general linear mixed model. Results: The lichen flora of the islets consisted of 326 taxa, the number of lichen species per islet varied from 2 to 197. Total number of species per islet and within the substrate groups was positively correlated with islet area and with number of biotopes, and negatively correlated with distance from the mainland; however, these relationships varied among the substrate groups. Although individual lichen species showed variation in responses, general trends in island biogeography were evident. Conclusions: The distribution pattern of lichens on the studied islets follows the theory of island biogeography: the number of species per islet depends on isolation, area and biotope diversity. Species specific traits, such as dispersal strategy and growth form, as well as availability of a particular substrate are important for formation of the lichen flora on islets. [source] Sources of variation in fecal cortisol levels in howler monkeys in belizeAMERICAN JOURNAL OF PRIMATOLOGY, Issue 7 2010Alison M. Behie Abstract High cortisol levels are known to cause low fecundity and increased mortality; thus, the prospect of using cortisol as a measure of population health is an exciting one. However, because so many factors can interact to influence cortisol release, it can be difficult to interpret what exactly is creating changes to cortisol levels. This study investigates variation in fecal cortisol levels in a population of black howlers (Alouatta pigra) from 350 fecal samples collected from 33 individuals in more than 4 years. A general linear mixed model revealed that cortisol varied significantly with fruit availability and contact with tourists. When fruit availability was low, cortisol increased, likely because when fruit availability is low monkeys eat less fruit, thus obtaining less sugar. This result may simply reflect cortisol's metabolic function of mobilizing glucose. It also indicates that these monkeys may be experiencing periods of food stress throughout the year, which was earlier thought to be minimal for a primarily folivorous species. Presence of tourists was the only other factor found to lead to high cortisol; with exposure to tourists increasing stress levels. These results highlight the importance of understanding how physiological factors can influence cortisol, making it easier to interpret results and determine the external social or ecological stressors that may increase cortisol. Am. J. Primatol. 72:600,606, 2010. © 2010 Wiley-Liss, Inc. [source] The importance of weather and agronomic factors for the overwinter survival of yellow rust (Puccinia striiformis) and subsequent disease risk in commercial wheat crops in EnglandANNALS OF APPLIED BIOLOGY, Issue 3 2007P. Gladders Abstract Disease survey data from 4475 randomly selected crops of wheat from England and Wales during 1985,2000 showed that yellow rust was most prevalent in 1988, 1989, 1990, 1998 and 1999. Disease severity on the upper two leaves was low as >95% crops had received foliar fungicides. Factors affecting the presence or absence (incidence) of yellow rust were investigated using random effects logistic regression (general linear mixed model). This enabled crop management (risk) variables for individual crops to be combined with meteorological variables measured at the county level. Two models are presented that analysed the effect of host genotype on incidence either solely through yellow rust resistance rating (Model 1) or by including both resistance rating (fixed effect) and cultivar (fitted as a random term) (Model 2). In both models, the percentage of crops with yellow rust decreased with cultivar disease resistance ratings ,3, the occurrence of severe frosts (<,5°C), use of systemic seed treatment and application of foliar fungicide sprays. There were no significant effects (P < 0.05) of timing of fungicide sprays, previous cropping or summer weather. The use of risk variables associated with overwintering survival may help adjust fungicide inputs to seasonal risk. [source] Combined cell wall polysaccharide, mycotoxin and bacterial lipopolysaccharide exposure and inflammatory cytokine responsesAPMIS, Issue 7 2009LENE JOHANNESSEN Human exposure to environmental microbes occurs regularly. Microbial compounds may interact with each other to affect cellular responses. We hypothesized that interactions between microbial compounds could modulate inflammatory cytokine responses in vitro. We investigated monocyte production of the pro-inflammatory cytokine tumour necrosis factor-, (TNF-,) and the regulatory cytokine interleukin-10 (IL-10) after combined exposure to the fungal cell wall polysaccharide mannan and to the ,-glucan laminarin, the mycotoxin citrinin and bacterial lipopolysaccharide (LPS). Interactions between the cell wall microbial compounds were estimated statistically in a general linear mixed model. We found that LPS (100 ng/ml) and the used ,-glucan (up to 1000 ,g/ml) significantly interacted with each other to reduce TNF-, production. Mannan (up to 100 ,g/ml) did not interact with the ,-glucan, but interacted with LPS. IL-10 production was induced by LPS only. The mycotoxin citrinin did not induce cytokine production, but was toxic to the cells in a dose- and time-dependent manner. However, non-toxic doses of citrinin reduced LPS-induced IL-10 production while LPS-induced TNF-, production was not similarly reduced by citrinin. In conclusion, interactions between microbial compounds can modulate cellular inflammatory cytokine production and experimental investigations of one compound at a time could give misleading conclusions about these combined effects. [source] | |||||||||||||