Community Ecology
Volume/Issue:
Volume 10: Issue 2
The effect of prevalence and its interaction with sample size on the reliability of species distribution models
Authors:
A. Jiménez-Valverde The University of Kansas Natural History Museum and Biodiversity Research Center Lawrence Kansas 66045 USA

Search for other papers by A. Jiménez-Valverde in
Current site
Google Scholar
PubMed Close
,
J. Lobo Museo Nacional de Ciencias Naturales Dpto. Biodiversidad y Biología Evolutiva. C/ José Gutiérrez Abascal 2 E-28006 Madrid Spain

Search for other papers by J. Lobo in
Current site
Google Scholar
PubMed Close
, and
J. Hortal Imperial College London NERC Centre for Population Biology, Division of Biology Silwood Park Campus, Ascot Berkshire SL5 7PY UK

Search for other papers by J. Hortal in
Current site
Google Scholar
PubMed Close
Pages:
196–205
Online Publication Date:
21 Dec 2009
Publication Date:
01 Dec 2009
Article Category:
Journal Article
DOI:
https://doi.org/10.1556/comec.10.2009.2.9
Keywords:
Logistic regression; Model accuracy; Prevalence; Sample size; Species distribution modelling
Restricted access

Prevalence (the presence/absence ratio in the training data) is commonly thought to influence the reliability of the predictions of species distribution models. However, little is known about its precise impact. We studied its effects using a virtual species, avoiding the presence of unaccounted-for effects in the modeling process (false absences, non-explanatory predictors, etc.). We sampled the distribution of the virtual species to obtain several data subsets of varying sample size and prevalence, and then modeled these data subsets using logistic regressions. Our results show that model predictions can be highly accurate over a wide range of sample sizes and prevalence scores, provided that the predictors are truly related to the distribution of the species and the training data are reliable. The effect of sample size becomes apparent for datasets of less than 70 data points, and the effect of prevalence is significant only for datasets with extremely unbalanced samples (<0.01 and >0.99). There is also a strong interaction between sample size and prevalence, indicating that the most negative factor is the sample size of each event (absence and/or presence), and not biased prevalence, as previously thought. We suggest that, in the real world, an interaction must exist between the sample size of each event and the quality of the training data. We discuss that biased prevalences can be a desirable property of the data, instead of a problem to be avoided, also pointing out the importance of using the best absence data possible when modeling the distribution of species of narrow geographic range.

  • Collapse
  • Expand
Cover Community Ecology
Community Ecology An Interdisciplinary Journal Reporting Progress in Community and Population Studies
Print ISSN:
1585-8553
Online ISSN:
1588-2756

To see the editorial board, please visit the website of Springer Nature.

Manuscript Submission: HERE

For subscription options, please visit the website of Springer Nature.

Community Ecology
Language English
Size A4
Year of
Foundation		 2000
Volumes
per Year		 1
Issues
per Year		 3
Founder Akadémiai Kiadó
Founder's
Address		 H-1117 Budapest, Hungary 1516 Budapest, PO Box 245
Publisher Akadémiai Kiadó
Springer Nature Switzerland AG
Publisher's
Address		 H-1117 Budapest, Hungary 1516 Budapest, PO Box 245.
CH-6330 Cham, Switzerland Gewerbestrasse 11.
Responsible
Publisher		 Chief Executive Officer, Akadémiai Kiadó
ISSN 1585-8553 (Print)
ISSN 1588-2756 (Online)

Monthly Content Usage

Abstract Views Full Text Views PDF Downloads
Sep 2024 34 1 1
Oct 2024 54 0 0
Nov 2024 54 0 0
Dec 2024 31 0 0
Jan 2025 45 2 2
Feb 2025 25 0 0
Mar 2025 0 0 0