Explore the vast range of titles available.

In ecology and evolution, meta‐analysis is an important tool to synthesise findings across separate studies and identify sources of heterogeneity. However, ecological and evolutionary data often exhibit complex dependence structures, such as shared sources of variation within studies, phylogenetic relationships and hierarchical sampling designs. Recent statistical advancements offer approaches for handling such complexities in dependence, yet these methods remain under‐utilised or unfamiliar to ecologists and evolutionary biologists. We conducted extensive simulations to evaluate modelling approaches for handling dependence in effect sizes and sampling errors in ecological and evolutionary meta‐analyses. We assessed the performance of multilevel models, incorporating an assumed sampling error variance–covariance (VCV) matrix (which account for within‐study correlation), cluster robust variance estimation (CRVE) methods and their combination across different true within‐study correlations. Finally, we showcased the applications of these models in two case studies of published meta‐analyses. Multilevel models produced unbiased regression coefficient estimates, and when a sampling VCV matrix was used, it provided accurate random effect variance components estimates within and among studies. However, the latter had no impact on regression coefficient estimates if the model was misspecified. In simulations involving phylogenetic multilevel meta‐analysis, models using CRVE methods generated narrower confidence intervals and lower coverage rates than the nominal expectations. The case study results showed the importance of considering a sampling error VCV matrix to improve the model fit. Our results provide clear modelling recommendations for ecologists and evolutionary biologists conducting meta‐analyses. To improve the precision of variance component estimates, we recommend constructing a VCV matrix that accounts for dependencies in sampling errors within studies. Although CRVE methods provide robust inference under certain conditions, we caution against their use with crossed random effects, such as phylogenetic multilevel meta‐analyses, as CRVE methods currently do not account for multi‐way clustering and may inflate Type I error rates. Finally, we recommend using multilevel meta‐analytic models to account for heterogeneity at all relevant hierarchical levels and to follow guidance on inference methods to ensure accurate coverage of the overall mean.

Aim Distribution of species is one of the most elementary but fundamental biodiversity patterns, yet too many puzzles remain unsolved. In most cases, distribution of species is not random, but presents some degree of aggregation. Accordingly, the term ‘distributional aggregation’ is prevailingly used in ecology and evolutionary biology to reflect the non‐random distribution characteristic of species in space and time. However, its meanings are multiform and can be decomposed into a variety of components. Location Global. Methods In this paper, through synthesizing historical literature and comparing relevant meanings of distributional aggregation under different contexts, we summarize the current statistical metrics in detecting and evaluating distributional aggregation that are suitable for different field‐survey methods, study models and sampling scales. In particular, we explore the concept under the multi‐species setting for which few conceptual advances and statistical methods have been developed. Results We propose pure data dispersion and spatiotemporal proximity, as two basic components of distributional aggregation. We further explore three advanced components of distributional aggregation: orthogonal, hierarchical and parallel components that can simultaneously link sampling taxa, sampling sites and sampling methods. Main Conclusions We hope the systematic review can serve as a potentially useful primer to ecologists for better understanding and investigating complex and new distributional patterns of biological diversity. We further provide informative guides on developing new statistical methods and metrics. We also discuss useful simulation algorithms for generating diverse distributional aggregation patterns of species, aiming to help ecologists to test and compare the performance of different metrics related to diversity and distribution patterns of species.

Meta-analysis is a statistical procedure for analyzing the combined data from different studies, and can be a major source of concise up-to-date information. The overall conclusions of a meta-analysis, however, depend heavily on the quality of the meta-analytic process, and an appropriate evaluation of the quality of meta-analysis (meta-evaluation) can be challenging. We outline ten questions biologists can ask to critically appraise a meta-analysis. These questions could also act as simple and accessible guidelines for the authors of meta-analyses. We focus on meta-analyses using non-human species, which we term ‘biological’ meta-analysis. Our ten questions are aimed at enabling a biologist to evaluate whether a biological meta-analysis embodies ‘mega-enlightenment’, a ‘mega-mistake’, or something in between.

Some capture–recapture models for population estimation cannot easily be fitted by the usual methods (maximum likelihood and Markov‐chain Monte Carlo). For example, there is no straightforward probability model for the capture of animals in traps that hold a maximum of one individual (‘single‐catch traps’), yet such data are commonly collected. It is usual to ignore the limit on individuals per trap and analyse with a competing‐risk ‘multi‐catch’ model that gives unbiased estimates of average density. However, that approach breaks down for models with varying density. Simulation and inverse prediction was suggested by Efford (2004) for estimating population density with data from single‐catch traps, but the method has been little used, in part because the existing software allows only a narrow range of models. I describe a new R package that refines the method and extends it to include models with varying density, trap interference and other sources of non‐independence among detection histories. The method depends on (i) a function of the data that generates a proxy for each parameter of interest and (ii) functions to simulate new datasets given values of the parameters. By simulating many datasets, it is possible to infer the relationship between proxies and parameters and, by inverting that relationship, to estimate the parameters from the observed data. The method is applied to data from a trapping study of brushtail possums Trichosurus vulpecula in New Zealand. A feature of these data is the high frequency of non‐capture events that disabled traps (interference). Allowing for a time‐varying interference process in a model fitted by simulation and inverse prediction increased the steepness of inferred year‐on‐year population decline. Drawbacks and possible extensions of the method are discussed.

One of the major benefits of interdisciplinary research is the chance to swap tools between fields, to save having to reinvent the wheel. The fields of language evolution and evolutionary biology have been swapping tools for centuries to the enrichment of both. Here I will discuss three categories of tool swapping: (1) conceptual tools, where analogies are drawn between hypotheses, patterns or processes, so that one field can take advantage of the path cut through the intellectual jungle by the other; (2) theoretical tools, where the machinery developed to process the data in one field is adapted to be applied to the data of the other; and (3) analytical tools, where common problems encountered in both fields can be solved using useful tricks developed by one or the other. I will argue that conceptual tools borrowed from linguistics contributed to the Darwinian revolution in biology; that theoretical tools of evolutionary change can in some cases be applied to both genetic and linguistic data without having to assume the underlying evolutionary processes are exactly the same; and that there are practical problems that have long been recognised in historical linguistics that may be solved by borrowing some useful analytical tools from evolutionary biology.

Conducting biodiversity surveys using a fully randomised design can be difficult due to budgetary constraints ( e.g. , the cost of labour), site accessibility, and other constraints. To this end, ecologists usually select representative line transects or quadrats from a studied area to collect individuals of a given species and use this information to estimate the levels of biodiversity over an entire region. However, commonly used biodiversity estimators such as Rao’s quadratic diversity index (and especially the Gini–Simpson index) were developed based on the assumption of independent sampling of individuals. Therefore, their performance can be compromised or even misleading when applied to species abundance datasets that are collected from non-independent sampling. In this study, we utilise a Markov chain model and derive an associated parameter estimator to account for non-independence in sequential sampling. Empirical tests on two forest plots in tropical (Barro Colorado, Island of Panama) and subtropical (Heishiding Nature Reserve of Guangdong, China) regions and the continental-scale spatial distribution of Acacia species in Australia showed that our estimators performed reasonably well. The estimated parameter measuring the degree of non-independence of subsequent sampling showed that a non-independent effect is very likely to occur when using line transects to sample organisms in subtropical regions at both local and regional spatial scales. In summary, based on a first-order Markov sampling model and using Rao’s quadratic diversity index as an example, our study provides an improvement in diversity estimation while simultaneously accounting for the non-independence of sampling in field biodiversity surveys. Our study presents one possible solution for addressing the non-independent sampling of individuals in biodiversity surveys.

Many recent evolutionary psychology and human behavioral ecology studies have tested hypotheses by examining correlations between variables measured at a group level (e.g., state, country, continent). In such analyses, variables collected for each aggregation are often taken to be representative of the individuals present within them, and relationships between such variables are presumed to reflect individual-level processes. There are multiple reasons to exercise caution when doing so, including: (1) the ecological fallacy, whereby relationships observed at the aggregate level do not accurately represent individual-level processes; (2) non-independence of data points, which violates assumptions of the inferential techniques used in null hypothesis testing; and (3) cross-cultural non-equivalence of measurement (differences in construct validity between groups). We provide examples of how each of these gives rise to problems in the context of testing evolutionary hypotheses about human behavior, and we offer some suggestions for future research.

Wildlife populations can be unmarked, meaning individuals lack distinguishing features for individual identification. Populations may also exhibit non‐independent movements, meaning individuals move together. For populations of either unmarked or non‐independent individuals, models based on spatial capture–recapture (SCR) approaches can be used to estimate abundance, density, and other parameters critical for monitoring, management, and conservation. However, when individuals are both unmarked and non‐independent, few model options are available. One approach has been to apply unmarked models and not address the non‐independence despite unquantified impacts on bias, precision, and the ability to make robust ecological inferences. We conducted a simulation study to quantify the impact of non‐independence on the performance of spatial count (SC) and spatial partial identity models (SPIM) – two SCR‐based unmarked modeling approaches – and used the performance of fully marked and independent SCR as a reference. We varied the levels of non‐independence (aggregation and cohesion), detection probability, and the number of partial identity covariates used to resolve identities in SPIM estimation. We expected abundance estimates to be increasingly biased and precise as aggregation and cohesion increased. Results showed that models indeed became less robust to increasing non‐independence, but importantly suggested that only SPIM could be reliably applied under low levels of cohesion when sufficient partial identity covariates are available. SC yielded consistently biased estimates with poor precision. SCR was consistently robust across combinations of aggregation and cohesion, as expected. We therefore advise against the use of SC models for estimating population parameters when individuals are known to be non‐independent, caution that SPIM may be used under narrow ecological conditions, and encourage continued investigations into sampling design and methods development for estimating populations of unmarked and non‐independent individuals.

Many important and interesting hypotheses about cultural evolution are evaluated using cross-cultural correlations: if knowing one particular feature of a culture (e.g. environmental conditions such as temperature, humidity or parasite load) allows you to predict other features (e.g. language features, religious beliefs, cuisine), it is often interpreted as indicating a causal link between the two (e.g. hotter climates carry greater disease risk, which encourages belief in supernatural forces and favours the use of antimicrobial ingredients in food preparation; dry climates make the production of distinct tones more difficult). However, testing such hypotheses from cross-cultural comparisons requires us to take proximity of cultures into account: nearby cultures share many aspects of their environment and are more likely to be similar in many culturally inherited traits. This can generate indirect associations between environment and culture which could be misinterpreted as signals of a direct causal link. Evaluating examples of cross-cultural correlations from the literature, we show that significant correlations interpreted as causal relationships can often be explained as a result of similarity between neighbouring cultures. We discuss some strategies for sorting the explanatory wheat from the co-varying chaff, distinguishing incidental correlations from causal relationships.

The value of meta-analysis in ecology hinges on the reproducibility of patterns generated by quantitative synthesis. Meta-analysts will vary in the criteria they use to screen studies and select data within studies, even when addressing exactly the same question. We summarize some of the many decisions that an ecologist must make in deciding what data to include in a synthesis. We then show, using multiple meta-analyses taken from the same literature on stream predation experiments, that meta-analytic conclusions can be colored by selection criteria that are not specifically a function of the relevance of the data. As a consequence, we recommend that meta-analysts perform several meta-analyses using different selection criteria to examine the robustness of reported findings. We also advise ecological meta-analysts to minimize use of selection criteria that are based on judgments of study quality when extracting data from the literature, because of the potential for unconscious bias. The influence of quality criteria on patterns in the data set can then be examined empirically. Our comparisons of mean effect size, for studies included vs. excluded on the basis of \"quality\" criteria, provided no evidence that rejected studies were aberrant or more variable than \"acceptable\" studies. One result of excluding such studies was a loss of statistical power. We urge ecologists to be more explicit about how data are selected for a meta-analysis, to examine the robustness of the patterns they report, and to conduct meta-analyses to describe as well as to infer.