Multiple Sources Of Variation Research Articles

Smoothed dynamic factor analysis for identifying trends in multivariate time series

Abstract Ecological processes are rarely directly observable, and inference often relies on estimating hidden or latent processes. State‐space models have become widely used for this task because of their ability to simultaneously estimate the multiple sources of variation (natural variability and variance attributed to observation errors). For multivariate time series, a second aim is often dimension reduction, or estimating a number of latent processes that are smaller than the number of observed time series. Dynamic factor analysis (DFA) has been used for performing time‐series dimension reduction, where latent processes are modelled as random walks. Whereas this may be suitable for some situations, random walks may be too flexible for other cases. Here, we introduce a new class of models, where latent processes are modelled as smooth functions (basis splines, penalized splines or Gaussian process models). We implement these models in our bayesdfa r package, which uses the rstan package for fitting. After evaluating model performance with simulated data, we apply conventional models and our smooth trend models to two long‐term datasets from the west coast of the United States: (a) a 35‐year dataset of pelagic juvenile rockfishes and (b) a 39‐year dataset of fisheries catches. Our simulations demonstrate that models matching the underlying trend smoothness make better out‐of‐sample predictions, but this advantage diminishes with increasing levels of observation error. For both case studies, the best smooth trend models had higher predictive accuracy, and yielded more precise predictions, compared to the conventional approach. The smooth trend factor models introduced here offer a new approach for state‐space dimension reduction of multivariate time series. These flexible Bayesian models may be particularly useful for data that are clumped in time, for data with high signal to noise ratios and generally for data where the underlying trend is assumed to be relatively smooth.

Sparse linear mixed model selection via streamlined variational Bayes

Linear mixed models are a versatile statistical tool to study data by accounting for fixed effects and random effects from multiple sources of variability. In many situations, a large number of candidate fixed effects is available and it is of interest to select a parsimonious subset of those being effectively relevant for predicting the response variable. Variational approximations facilitate fast approximate Bayesian inference for the parameters of a variety of statistical models, including linear mixed models. However, for models having a high number of fixed or random effects, simple application of standard variational inference principles does not lead to fast approximate inference algorithms, due to the size of model design matrices and inefficient treatment of sparse matrix problems arising from the required approximating density parameters updates. We illustrate how recently developed streamlined variational inference procedures can be generalized to make fast and accurate inference for the parameters of linear mixed models with nested random effects and global-local priors for Bayesian fixed effects selection. Our variational inference algorithms achieve convergence to the same optima of their standard implementations, although with significantly lower computational effort, memory usage and time, especially for large numbers of random effects. Using simulated and real data examples, we assess the quality of automated procedures for fixed effects selection that are free from hyperparameters tuning and only rely upon variational posterior approximations. Moreover, we show high accuracy of variational approximations against model fitting via Markov Chain Monte Carlo sampling.

Flexible approaches for estimating partial eta squared in mixed-effects models with crossed random factors.

Mixed-effects models are frequently used in a variety of disciplines because they can appropriately specify multiple sources of variation. However, precisely because they distinguish between multiple sources of variation, it is difficult to specify a standardized effect size, such as η2. Several approaches to this problem have been proposed, but most do not address models with crossed random factors, and none allows for the range of data and models that researchers typically test. For example, no existing approach handles random slopes for a continuous predictor. We introduce several new, flexible approaches to estimating η2 in mixed-effect models with crossed random factors. We then conduct a simulation to assess new and old methods. We examine their respective strengths and weaknesses and offer recommendations for a simple approach based on the work of Snijders and Bosker (2011).

ACE: a probabilistic model for characterizing gene-level essentiality in CRISPR screens

High-throughput CRISPR-Cas9 knockout screens are widely used to evaluate gene essentiality in cancer research. Here we introduce a probabilistic modeling framework, Analysis of CRISPR-based Essentiality (ACE), that accounts for multiple sources of variation in CRISPR-Cas9 screens and enables new statistical tests for essentiality. We show using simulations that ACE is effective at predicting both absolute and differential essentiality. When applied to publicly available data, ACE identifies known and novel candidates for genotype-specific essentiality, including RNA m6-A methyltransferases that exhibit enhanced essentiality in the presence of inactivating TP53 mutations. ACE provides a robust framework for identifying genes responsive to subtype-specific therapeutic targeting.

Potential impacts of projected warming scenarios on winter wheat in the UK

AbstractThe goals of this study are to analyse the impacts of 1.5 and 2.0°C scenarios on UK winter wheat using a combination of Global Climate Models (GCMs), crop models, planting dates and cultivars; to evaluate the impact of increased air temperature on winter wheat phenology and potential yield; to quantify the underlying uncertainties due to the multiple sources of variability introduced by climate scenarios, crop models and agronomic management. The data used to calibrate and evaluate three crop models were obtained from a field experiment with two irrigation amounts and two wheat cultivars that have different phenology and growth habit. After calibration, the model was applied to fifty locations across the wheat-growing area of the UK to cover all the main growing regions, with most points located in the main growing areas. Four GCMs, with two cultivars and five planting dates, were simulated at the end of the century. Results of this study showed that the UK potential wheat yield will increase by 2–8% under projected climate. Farmers will need to close such a gap in the future because it will have implications in terms of food security. Future climatic conditions might increase such a gap. Adaptation measures (e.g. moving the optimal planting time), along with climate-ready varieties bred for future conditions and with precision agriculture techniques can help to reduce this gap and ensure that the future actual UK wheat production will be close to the potential.

Reliable Estimationof CD8 T Cell Inhibition of In Vitro HIV-1 Replication.

The HIV-1 viral inhibition assay (VIA) measures CD8 T cell-mediated inhibition of HIV replication in CD4 T cells and is increasingly used for clinical testing of HIV vaccines and immunotherapies. The VIA has multiple sources of variability arising from in vitro HIV infection and co-culture of two T cell populations. Here, we describe multiple modifications to a 7-day VIA protocol, the most impactful being the introduction of independent replicate cultures for both HIV infected-CD4 (HIV-CD4) and HIV-CD4:CD8 T cell cultures. Virus inhibition was quantified using a ratio of weighted averages of p24+ cells in replicate cultures and the corresponding 95% confidence interval. An Excel template is provided to facilitate calculations. Virus inhibition was higher in people living with HIV suppressed on antiretroviral therapy (n=14, mean: 40.0%, median: 43.8%, range: 8.2 to 73.3%; p < 0.0001, two-tailed, exact Mann-Whitney test) compared to HIV-seronegative donors (n = 21, mean: -13.7%, median: -14.4%, range: -49.9 to 20.9%) and was stable over time (n = 6, mean %COV 9.4%, range 0.9 to 17.3%). Cross-sectional data were used to define 8% inhibition as the threshold to confidently detect specific CD8 T cell activity and determine the minimum number of culture replicates and p24+ cells needed to have 90% statistical power to detect this threshold. Last, we note that, in HIV seronegative donors, the addition of CD8 T cells to HIV infected CD4 T cells consistently increased HIV replication, though the level of increase varied markedly between donors. This co-culture effect may contribute to the weak correlations observed between CD8 T cell VIA and other measures of HIV-specific CD8 T cell function.

Variance components of ratings of physician-patient communication: A generalizability theory analysis.

BackgroundThe ratings of physician-patient communication are an important indicator of the quality of health care delivery and provide guidance for many important decisions in the health care setting and in health research. But there is no gold standard to assess physician-patient communication. Thus, depending on the specific measurement condition, multiple sources of variance may contribute to the total score variance of ratings of physician-patient communication. This may systematically impair the validity of conclusions drawn from rating data.ObjectiveTo examine the extent to which different measurement conditions and rater perspectives, respectively contribute to the variance of physician-patient communication ratings.MethodsVariance components of ratings of physician-patient communication gained from 32 general practitioners and 252 patients from 25 family practices in Germany were analyzed using generalizability theory. The communication dimensions “shared decision making”, “effective and open communication” and “satisfaction” were considered.ResultsPhysician-patient communication ratings most substantially reflect unique rater-perspective and communication dimension combinations (32.7% interaction effect). The ratings also represented unique physician and rater-perspective combinations (16.3% interaction effect). However, physicians’ communication behavior and the observed communication dimensions revealed only a low extent of score variance (1% physician effect; 3.7% communication dimension effect). Approximately half of the variance remained unexplained (46.2% three-way interaction, confounded with error).ConclusionThe ratings of physician-patient communication minimally reflect physician communication skills in general. Instead, these ratings exhibit primarily differences among physicians and patients in their tendency to perceive shared decision making and effective and open communication and to be satisfied with communication, regardless of the communication behavior of physicians. Rater training and assessing low inferential ratings of physician-patient communication dimensions should be considered when subjective aspects of rater perspectives are not of interest.

An Effectiveness Study of SPC Control Charts on the Quality Control for Construction in Asphalt Pavement

Aiming at the problems of insufficient dynamic control and data mining of Expressway pavement construction quality in Yunnan, this paper analyzes the application of Exponentially Weighted Moving Average (EWMA) control chart in dynamic control of Expressway asphalt pavement construction quality. Firstly, the quality characteristic value of pavement construction is defined as the data type of multiple variation sources, and the mathematical model of multiple variation sources is studied to solve the problem of excessive false alarm. Then, on the theoretical basis of statistical process control (SPC), the paper compares and analyzes the three control chart schemes of traditional Hugh Hart control chart, Tabular Cumulative Sum (CUSUM) control chart and EWMA control chart, and concludes that the application effect of EWMA control chart is obviously better than the other methods in the aspects of applicability and inspection output. Finally, combining with the measured data of asphalt mixture temperature and asphalt weight in pavement construction, EWMA control diagram is used to carry out practical research, and the results are better.

Individual variation in thermal plasticity and its impact on mass‐scaling

Physiological processes vary widely across individuals and can influence how individuals respond to environmental change. Repeatability in how metabolic rate changes across temperatures (i.e. metabolic thermal plasticity) can influence mass‐scaling exponents in different thermal environments. Moreover, repeatable plastic responses are necessary for reaction norms to respond to selective forces which is important for populations living in fluctuating environments. Nonetheless, only a small number of studies have explicitly quantified repeatability in metabolic plasticity, and fewer have explored how it can impact mass‐scaling. We repeatedly measured standard metabolic rate of n = 42 delicate skinks Lampropholis delicata at six temperatures over the course of four months (N[observations] = 4952). Using hierarchical statistical techniques, we accounted for multi‐level variation and measurement error in our data in order to obtain more precise estimates of reaction norm repeatability and mass‐scaling exponents at different acute temperatures. Our results show that individual differences in metabolic thermal plasticity were somewhat consistent over time (Rslope = 0.25, 95% CI = 2.48 × 10−8 – 0.67), however estimates were associated with a large degree of error. After accounting for measurement error, which decreased steadily with temperature, we show that among individual variance remained consistent across all temperatures. Congruently, temperature specific repeatability of average metabolic rate was stable across temperatures. Cross‐temperature correlations were positive but were not uniform across the reaction norm. After taking into account multiple sources of variation, our estimates for mass‐scaling did not change with temperature and were in line with published values for snakes and lizards. This implies that repeatable plastic responses may promote thermal stability of scaling exponents. Our work contributes to understanding how energy expenditure scales with abiotic and biotic factors and the capacity for reaction norms to respond to selection.

Causes of plasticity and consistency of dispersal behaviour in a group-living spider

Within animal populations, behaviour often varies consistently among individuals, yet it remains unclear what generates and maintains this variation in ‘animal personality’. To understand how multiple sources of variation might combine to promote interindividual differences in behaviour, we examined dispersal behaviour in a colonial spider, Cyrtophora citricola. We measured the repeatability of two behaviours in the laboratory that are associated with dispersal: latency to tiptoe (a dispersal behaviour in spiders) and latency to explore a new environment. We then tested five hypothesized sources of consistent variation across two generations of spiders: sex, food resources, transient internal states, heritability and parental effects. Spiders were modestly consistent in both latency to tiptoe and latency to explore. Males behaved similarly to females of the same size; however, males mature at a smaller size than females, and females became much more hesitant to tiptoe and explore as they grew. Food availability had no effect on either behaviour. Repeatability of exploration was higher within an instar than across instars, suggesting that some aspect of spiders' internal state, other than food, maintained relatively high repeatability over short periods. Tiptoe behaviour was significantly heritable for female relatives, but not for male relatives. For exploratory behaviour, siblings showed significant heritability, while parents and offspring did not, suggesting that early natal environments shared by siblings, rather than genes, might account for their similar behaviour. Finally, although food had no effect on the generation of spiders to which the diet was applied, poorly fed fathers had offspring that tiptoed more readily than well-fed fathers, suggesting paternal effects potentially transmitted through epigenetics. Tiptoe and exploratory behaviours were correlated within individuals, yet the overall modest interindividual variation in these two dispersal behaviours was maintained by disparate effects of different sources of variation.

Variation in cross-sectional shape and biomechanical properties of the bat humerus under Wolff's law.

Bats use their forelimbs in different ways, but flight is the most notable example of morphological adaptation. Foraging and roosting specializations beyond flight have also been described in several bat lineages. Understanding postcranial evolution during the locomotory and foraging diversification of bats is fundamental to understanding bat evolution. We investigated whether different foraging and roosting behaviors influenced humeral cross-sectional shape and biomechanical variation, following Wolff's law of bone remodeling. The effect of body size and phylogenetic relatedness was also tested, in order to evaluate multiple sources of variation. Our results suggest strong ecological signal and no phylogenetic structuring in shape and biomechanical variation in humeral phenotypes. Decoupled modes of scaling of shape and biomechanical variation were consistently indicated across foraging and roosting behaviors, suggesting divergent allometric trajectories. Terrestrial locomoting and upstand roosting species showed unique patterns of shape and biomechanical variation across all our analyses, suggesting that these rare behaviors among bats place unique functional demands on the humerus, canalizing phenotypes. Our results suggest that complex and multiple adaptive pathways interplay in the postcranium, leading to the decoupling of different features and regions of skeletal elements optimized for different functional demands. Moreover, our results shed further light on the phenotypical diversification of the wing in bats and how adaptations besides flight could have shaped the evolution of the bat postcranium.

Political Ideology Modifies the Effect of Glass Cliff Candidacies on Election Outcomes for Women in American State Legislative Races (2011-2016).

Research on glass cliff political candidacies shows that compared to men, women are more likely to run for office in districts where they are likely to lose. We examined if party differences in whether female candidates face these worse conditions in the United States could account for persistent and growing party and state variation in women’s representation. Using election data from 2011 to 2016, we compared Republican versus Democratic candidacies at the state legislative level. We found that women in both parties faced glass cliffs in House races, but not in the Senate. For Republican women, glass cliff conditions accounted for worse election outcomes, but Democratic women were more likely to win when these conditions were considered. Variation in party by state measures of glass cliff effects were also found to explain state variation in women’s office holding. We found that for Democrats, more women win when more women run, but for Republicans, more women win only when the seats they face are more winnable. These results point to the role of polarized traditional versus progressive political ideologies in structuring the motives which underlie glass cliff conditions for women in politics, suggesting that practical solutions be tailored to party. To overcome the growing gap in women’s representation, current efforts to increase the quantity of women running would be complemented by a focus on improving the quality of contests they face, with Republican women most likely to benefit. Further research attending to the multiple sources of variation which impact gendered election outcomes can inform more targeted solutions for advancing equality. Online slides for instructors who want to use this article for teaching are available on PWQ's website at http://journals.sagepub.com/doi/suppl/10.1177/0361684321992046

A new tool for technical standardization of the Ki67 immunohistochemical assay.

Ki67, a nuclear proliferation-related protein, is heavily used in anatomic pathology but has not become a companion diagnostic or a standard-of-care biomarker due to analytic variability in both assay protocols and interpretation. The International Ki67 Working Group in breast cancer has published and has ongoing efforts in the standardization of the interpretation of Ki67, but they have not yet assessed technical issues of assay production representing multiple sources of variation, including antibody clones, antibody formats, staining platforms, and operators. The goal of this work is to address these issues with a new standardization tool. We have developed a cell line microarray system in which mixes of human Karpas 299 or Jurkat cells (Ki67+) with Sf9 (Spodoptera frugiperda) (Ki67-) cells are present in incremental standardized ratios. To validate the tool, six different antibodies, including both ready-to-use and concentrate formats from six vendors, were used to measure Ki67 proliferation indices using IHC protocols for manual (bench-top) and automated platforms. The assays were performed by three different laboratories at Yale and analyzed using two image analysis software packages, including QuPath and Visiopharm. Results showed statistically significant differences in Ki67 reactivity between each antibody clone. However, subsets of Ki67 assays using three clones performed in three different labs show no significant differences. This work shows the need for analytic standardization of the Ki67 assay and provides a new tool to do so. We show here how a cell line standardization system can be used to normalize the staining variability in proliferation indices between different antibody clones in a triple negative breast cancer cohort. We believe that this cell line standardization array has the potential to improve reproducibility among Ki67 assays and laboratories, which is critical for establishing Ki67 as a standard-of-care assay.

Taxing Property in Developing Countries: Theory and Evidence from Mexico

Property taxes in developing countries are plagued by noncompliance and can exacerbate liquidity constraints. We characterize optimal enforcement and taxation policies as functions of revenue elasticities and measures of taxpayer hardship. We estimate these parameters using multiple sources of variation and administrative data from Mexico City. Both rate increases and enhanced enforcement raise revenue, but liquidity constraints also shape taxpayer behavior. Despite the presence of liquidity constraints, we find that raising tax rates increases welfare. In contrast, enforcement generates higher private costs than welfare benefits. On the margin, welfare-maximizing governments would prefer to increase tax rates rather than enhance enforcement.

Hierarchical Bayesian approaches to statistical modelling of geotechnical data

ABSTRACT Geotechnical data used for reliability-based design (RBD) and load and resistance factor design (LRFD) calibration can be parsed into subgroups based on material type, location, test method, and so on. Most often statistical analyses assume all data fall within a representative envelope, and pool (combine) all data into a single large data set without rigorously evaluating the veracity of this assumption. Adopting a Bayesian view, we introduce hierarchical/multilevel models as the more suitable alternative that takes into account multiple variation sources. Rather than assuming an identical parameter for data in all groups, hierarchical models assume exchangeable group-specific parameters, or informally, “similar but not identical” parameters. The utility of Bayesian hierarchical modelling is demonstrated by examining the accuracy of three reinforcement load models for polyester strap mechanically stabilised earth (MSE) walls and a simple hierarchical model as an example. We show that hierarchical models (i) have the customary complete pooling approach as their limiting case, (ii) quantify uncertainty from different sources of variation, (iii) prevent under-fitting, and (iv) can be used as tools for understanding the variations in the data. A simple RBD/LRFD example shows practical implications of hierarchical modelling of bias data. Finally, more complex/flexible hierarchical models are briefly discussed. Abbreviations: AASHTO: American Association of State Highway and Transportation Officials; ANOVA: analysis of variance; CDF: cumulative distribution function; CI: credible interval; COV coefficient of variation; DIC: deviance information criterion; LOGOIC: leave-one-group-out information criteria; LOOIC: leave-one-out information criteria; LM: load model; lppd: log pointwise predictive density; LRFD: load and resistance factor design; MCMC: Markov chain Monte Carlo; MSE: mechanically stabilised earth (wall); NHT: null hypothesis testing; PET: polyester; PGM: probabilistic graphical model; RBD: reliability-based design; WAIC: Watanabe-Akaike information criterion.

Improving the reliability of conductive atomic force microscopy-based electrical contact resistance measurements

Electrical contact resistance (ECR) measurements performed via conductive atomic force microscopy (C-AFM) suffer from poor reliability and reproducibility. These issues are due to a number of factors, including sample roughness, contamination via adsorbates, changes in environmental conditions such as humidity and temperature, as well as deformation of the tip apex caused by contact pressures and/or Joule heating. Consequently, ECR may vary dramatically from measurement to measurement even on a single sample tested with the same instrument. Here we present an approach aimed at improving the reliability of such measurements by addressing multiple sources of variability. In particular, we perform current-voltage spectroscopy on atomically flat terraces of highly oriented pyrolytic graphite (HOPG) under an inert nitrogen atmosphere and at controlled temperatures. The sample is annealed before the measurements to desorb adsorbates, and conductive diamond tips are used to limit tip apex deformation. These precautions lead to measured ECR values that follow a Gaussian distribution with significantly smaller standard deviation than those obtained under conventional measurement conditions. The key factor leading to this improvement is identified as the switch from ambient conditions to a dry nitrogen atmosphere. Despite these improvements, spontaneous changes in ECR are observed during measurements performed over several minutes. However, it is shown that such variations can be suppressed by applying a higher normal load.

Efficiency of scanning and attention to faces in infancy independently predict language development in a multiethnic and bilingual sample of 2-year-olds

Efficient visual exploration in infancy is essential for cognitive and language development. It allows infants to participate in social interactions by attending to faces and learning about objects of interest. Visual scanning of scenes depends on a number of factors, and early differences in efficiency are likely contributing to differences in learning and language development during subsequent years. Predicting language development in diverse samples is particularly challenging, as additional multiple sources of variability affect infant performance. In this study, we tested how the complexity of visual scanning in the presence or absence of a face at 6 to 7 months of age is related to language development at 2 years of age in a multiethnic and predominantly bilingual sample from diverse socioeconomic backgrounds. We used Recurrence Quantification Analysis to measure the temporal and spatial distribution of fixations recurring in the same area of a visual scene. We found that in the absence of a face the temporal distribution of re-fixations on selected objects of interest (but not all) significantly predicted both receptive and expressive language scores, explaining 16% to 20% of the variance. Also, lower rate of re-fixations recurring in the presence of a face predicted higher receptive language scores, suggesting larger vocabulary in infants that effectively disengage from faces. Altogether, our results suggest that dynamic measures, which quantify the complexity of visual scanning, can reliably and robustly predict language development in highly diverse samples. They suggest that selective attending to objects predicts language independently of attention to faces. As eye-tracking and language assessments were carried out in early intervention centres, our study demonstrates the utility of mobile eye-tracking setups for early detection of risk in attention and language development.

Augmented reality for next generation infrastructure inspections

This article introduces the use of emerging augmented reality technology to enable the next generation of structural infrastructure inspection and awareness. This work is driven by the prevalence of visual structural inspection. It is known that current visual inspection techniques have multiple sources of variance that should be reduced in order to achieve less ambiguous visual inspections. Emerging augmented reality tools feature a variety of sensors, computation, and communication resources that can enable relevant structural inspection data to be collected at very high resolution in an unambiguous manner. This work shows how emerging augmented reality tools can be used to greatly enhance our ability to capture comprehensive, high-resolution, three-dimensional measurements of critical infrastructure. This work also provides detailed information on the software architecture for augmented reality structural inspection applications that helps meet the goals of the framework. The fact that the framework is designed to accommodate the considerations associated with high-consequence infrastructure implies that it is also comprehensive enough to be applied to less hazardous but still high-value infrastructure such as bridges, dams, and tunnels. Augmented reality has great potential to enable the next generation of smart infrastructure, and this work focuses on addressing how augmented reality can be leveraged to enable the next generation of structural awareness for high-consequence, long-lifespan structures.

Non-Random Exposure to Exogenous Shocks: Theory and Applications

We develop new tools for estimating the causal effects of treatments or instruments that combine multiple sources of variation according to a known formula. Examples include treatments capturing spillovers in social and transportation networks, simulated instruments for policy eligibility, and shift-share instruments. We show how exogenous shocks to some, but not all, determinants of such variables can be leveraged while avoiding omitted variables bias. Our solution involves specifying counterfactual shocks that may as well have been realized and adjusting for a summary measure of non-randomness in shock exposure: the average treatment (or instrument) across such counterfactuals. We further show how to use shock counterfactuals for valid finite-sample inference, and characterize the valid instruments that are asymptotically efficient. We apply this framework to address bias when estimating employment effects of market access growth from Chinese high-speed rail construction, and to boost power when estimating coverage effects of expanded Medicaid eligibility.

Does predator abundance influence species diversity of equilibrium macroinvertebrate assemblages in spring fens?

Abstract Predation may significantly control number and density of coexisting species. The effects of predation on species diversity have traditionally been tested in experiments and theoretical models of simple trophic systems. In complex natural ecosystems, however, disentangling multiple sources of variation is difficult. In groundwater‐fed environments, a significant effect of predation can be expected due to the relatively stable environmental conditions; however, it has never been properly examined. We analysed species diversity and total abundance of macroinvertebrate assemblages in 48 Western Carpathian spring fens, separately for whole sites and mesohabitat/season, and partitioned the effects of predation intensity from those of environmental variables in robust models using a bootstrapping technique. We verified our results by accounting for taxa resistant to predation. The assumption that predation‐mediated coexistence of species is the main mechanism responsible for the relatively species‐rich assemblages in the Western Carpathian spring fens was not supported. However, predation may significantly influence abundance of non‐predatory species and, under some conditions, it may contribute to explaining patterns in species diversity. The effect of predation did not differ between the mesohabitats with different stability. However, we found higher environmental control in spring and a stronger effect of predators in autumn, which suggests that different mechanisms influence fen assemblages in different seasons. Our study provides a new robust approach how to test the effect of predation on natural macroinvertebrate assemblages. The importance of predation was lower than expected in equilibrium assemblages but it may vary in time.