Transcending data gaps: a framework to reduce inferential errors in ecological analyses

Kim, Sun Wook; Blomberg, Simon P.; Pandolfi, John M.

doi:10.1111/ele.13089

Cited by 39 publications

(53 citation statements)

References 90 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Traits were simulated under a Brownian model of evolution, with a Gaussian distribution of values ranging from zero to 10 to mimic the distribution of real trait data on a logarithmic scale (a transformation often used in comparative studies). The impact of phylogenetic signal strength on imputation performance was already tested by Kim et al (2018) and Molina‐Venegas et al (2018); therefore, we standardized Pagel’s λ between the phylogeny and traits at approximately one. The response was simulated as a product of a trait, rather than through the phylogeny, and has a Gaussian distribution ranging from zero to 10.…”

Section: Methodsmentioning

confidence: 99%

“…Therefore, correlations and phylogenetic information can be used to predict missing trait values more accurately (Penone et al, 2014; Swenson, 2014). Previous studies have suggested that imputation in ecological and evolutionary studies generally outperforms complete‐case analysis (Kim et al, 2018; Little & Rubin, 2002; Penone et al, 2014). However, imputation can only be successful if it accounts for the mechanism by which data are missing.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Handling missing values in trait data

Johnson

Isaac

Paviolo

et al. 2020

Global Ecol. Biogeogr.

119

122

View full text Add to dashboard Cite

Aim Trait data are widely used in ecological and evolutionary phylogenetic comparative studies, but often values are not available for all species of interest. Traditionally, researchers have excluded species without data from analyses, but estimation of missing values using imputation has been proposed as a better approach. However, imputation methods have largely been designed for randomly missing data, whereas trait data are often not missing at random (e.g., more data for bigger species). Here, we evaluate the performance of approaches for handling missing values when considering biased datasets. Location Any. Time period Any. Major taxa studied Any. Methods We simulated continuous traits and separate response variables to test the performance of nine imputation methods and complete‐case analysis (excluding missing values from the dataset) under biased missing data scenarios. We characterized performance by estimating the error in imputed trait values (deviation from the true value) and inferred trait–response relationships (deviation from the true relationship between a trait and response). Results Generally, Rphylopars imputation produced the most accurate estimate of missing values and best preserved the response–trait slope. However, estimates of missing data were still inaccurate, even with only 5% of values missing. Under severe biases, errors were high with every approach. Imputation was not always the best option, with complete‐case analysis frequently outperforming Mice imputation and, to a lesser degree, BHPMF imputation. Mice, a popular approach, performed poorly when the response variable was excluded from the imputation model. Main conclusions Imputation can handle missing data effectively in some conditions but is not always the best solution. None of the methods we tested could deal effectively with severe biases, which can be common in trait datasets. We recommend rigorous data checking for biases before and after imputation and propose variables that can assist researchers working with incomplete datasets to detect data biases and minimize errors.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Handling missing values in trait data

Johnson

Isaac

Paviolo

et al. 2020

Global Ecol. Biogeogr.

119

122

View full text Add to dashboard Cite

show abstract

“…To achieve complete species trait coverage, we imputed missing data using multivariate imputation with chained equations (MICE), based on functional (the transformed traits) and phylogenetic [the first 10 phylogenetic eigenvectors extracted from trees for birds (Prum et al, ) and mammals (Fritz, Bininda‐Emonds, & Purvis, )] relationships between species. MICE has been shown to have improved sample size and smaller error and bias than the data deletion approach and other multiple imputation approaches (Kim, Blomberg, & Pandolfi, ; Penone et al, ; Taugourdeau, Villerd, Plantureux, Huguenin‐Elie, & Amiaud, ). The data deletion approach was performed for comparative purposes (Supporting Information Appendix S4, Figure S4.1, Table S4.3).…”

Section: Methodsmentioning

confidence: 99%

Global trade‐offs of functional redundancy and functional dispersion for birds and mammals

Cooke

Bates

Eigenbrod

2019

Global Ecol Biogeogr

119

109

View full text Add to dashboard Cite

Aim The diversity of birds and mammals is typically described in separate analyses, but species may play similar roles. Here, we develop a comparative trait framework for birds and mammals to provide a global quantification of the similarity of species roles (functional redundancy) and the breadth of roles across taxa (functional dispersion). We predict different contributions of birds and mammals to redundancy and dispersion, and unique geographical patterns of redundancy and dispersion by including both taxa. Location Global. Time period Contemporary. Major taxa studied Birds and mammals. Methods We systematically select, compile and impute the same six traits (i.e., a common currency of traits) across 15,485 bird and mammal species from multiple databases. We use these six traits to compute functional redundancy and functional dispersion for birds and mammals across all 825 terrestrial ecoregions. We then calculate the standardized effect size (SES) of these observed values compared with null expectations, based on a randomization of species composition (i.e., independent of differences in species richness). Results We find that species‐rich regions, such as the Neotropics, have high functional redundancy coupled with low functional dispersion, characterizing a global trade‐off. Thus, in general, as species richness increases, the similarity in species functional roles also increases. We therefore suggest that different processes generate species richness/functional redundancy and functional dispersion, leading to a novel, and generally non‐tropical, distribution of hotspots of high functional dispersion across Madagascar, Eastern Asia and Western USA. Main conclusions We recommend consideration of both the similarity and the breadth of functional roles across species pools, including taxa that may play similar roles. We therefore suggest that functional redundancy, as a means of insurance, and functional dispersion, as an indicator of response diversity, should be evaluated further as conservation objectives.

show abstract

“…The first 10 eigenvectors in our data represented 61% of the variation in the phylogenetic distances among seabirds. Phylogenetic data can improve the estimation of missing trait values in the imputation process (Kim et al, 2018;Swenson, 2014). This is because closely related species tend to be more similar to each other (Pagel, 1999) and many traits display high degrees of phylogenetic signal (Blomberg, Garland, & Ives, 2003).…”

Section: Multiple Imputationmentioning

confidence: 99%

“…Blomberg, & Pandolfi, 2018; Penone et al, 2014;Taugourdeau, Villerd, Plantureux, Huguenin- Elie, & Amiaud, 2014).…”

mentioning

confidence: 99%

Biological traits of seabirds predict extinction risk and vulnerability to anthropogenic threats

Richards

Cooke

Bates

2020

Preprint

View full text Add to dashboard Cite

Aim: Here we aim to: 1) test whether globally-threatened vs non-threatened seabirds are separated in trait space; 2) quantify the redundancy and uniqueness of species trait combinations per IUCN Red List Category; and 3) identify traits that render species vulnerable to anthropogenic threats. Location: Global. Time period: Contemporary. Major taxa studied: Seabirds. Methods: We compile and impute eight traits that relate to species vulnerabilities and ecosystem functioning across 341 seabird species. Using these traits, we build a mixed data PCA of species trait space. We further quantify trait redundancy with a unique trait combinations (UTCs) approach. Finally, we employ a similarity of percentages analysis (SIMPER) to identify which traits explain the greatest difference between threat groups. Results: We find seabirds segregate in trait space based on threat status, indicating anthropogenic impacts are selectively removing large, long-lived, pelagic surface feeders with narrow habitat breadths. We further find that globally threatened species have higher trait redundancy, while non-threatened species have relatively unique ecological strategies and limited redundancy. Finally, we find that species with narrow habitat breadths, fast reproductive speeds, and omnivorous diets are more likely to be threatened by habitat-modifying processes (e.g., pollution and natural system modifications); whereas pelagic specialists with slow reproductive speeds and omnivorous diets are vulnerable to threats that directly impact survival and fecundity (e.g., invasive species and biological resource use). Main conclusions: Our results suggest both globally threatened and non-threatened species contribute unique ecological strategies. Consequently, conserving both threat groups, but with contrasting approaches may avoid potential changes in ecosystem functioning and stability.

show abstract

Transcending data gaps: a framework to reduce inferential errors in ecological analyses

Cited by 39 publications

References 90 publications

Handling missing values in trait data

Handling missing values in trait data

Global trade‐offs of functional redundancy and functional dispersion for birds and mammals

Biological traits of seabirds predict extinction risk and vulnerability to anthropogenic threats

Contact Info

Product

Resources

About