Bias and information in biological records

By Nick J B Isaac, Michael J. O. Pocock. 2015

345 citations recorded by the source corpus.

Subfields: Epidemiology, Public Health, Environmental and Occupational Health, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/4ad8501504f6da8accfad94114672c969e1ca8a6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Epidemiology

G. Guillera‐Arroita

In Statistics and Probability

Citations in the corpus

The Biological Records Centre: a pioneer of citizen science (2015, background): “Coverage is patchy for some of the less well recorded taxa (Fig 1C; Isaac & Pocock, 2015; Preston & Pearman, 2015), but gaps are often explicitly targeted with additional recording.”
Unstructured citizen science data fail to detect long‐term population declines of common birds in Denmark (2016, methodology): “Using data sets with higher information content that allow the statistical modelling of detection probability can yield improved trend estimates (K ery et al., 2010a; Isaac & Pocock, 2015), but whether such filtering can overcome other deficiencies of unstructured citizen science data is unclear.”
Interpreting insect declines: seven challenges and a way forward (2020, methodology): “Here, we provide only a partial (and cursory) set of examples of detection bias (for further examples, see Isaac & Pocock, 2015), and there are likely to be many other situations in which temporal autocorrelation in detectability might occur (setting aside the apocryphal car windscreen design…”
Monitoring insect pollinators and flower visitation: The effectiveness and feasibility of different survey methods (2019, methodology): “These data are collected using unstandardized or semi-standardized protocols (Isaac & Pocock, 2015) and changes in recording intensity, taxonomic ability and sampling strategies mean sources of bias have not been consistent over time.”
Data Integration for Large-Scale Models of Species Distributions. (2020, background): “Spatial biases are particularly widespread and problematic in presence-only data [26,27], and these generally need to be accounted for in the observation model in order to estimate the true habitat preferences.”
Sampling biases shape our view of the natural world (2020, background): “We do not seek to dismiss the use of citizen science data for understanding biological patterns (Amano et al. 2016), and although biases that can exist in such data have been acknowledged (Isaac and Pocock 2015), its impact on global diversity mapping is less well described.”