Checkerboard score

This article relies largely or entirely on a single source. Relevant discussion may be found on the talk page. Please help improve this article by introducing citations to additional sources. Find sources: "Checkerboard score" – news · newspapers · books · scholar · JSTOR (September 2013)

In biodiversity studies, the checkerboard score or C-score is a statistic which determines the randomness of the distribution of two or more species through a collection of biomes. The statistic, first published by Stone and Roberts in 1990,^[1] expands on the earlier work of Diamond^[2] that defined a notion of "checkerboard distributions" as an indicator of species competition.

A low c-score indicates a higher randomness, i.e. a greater likelihood that the distribution of one species has not been directly affected by the presence of other species.

Given two species sp₁, sp₂ and n islands, an incident matrix is built. In the ${\displaystyle 2\times n}$ incident matrix, each row represents one of the two species and each column represents a different island. The matrix is then filled with each cell being set to either 0 or 1. Cell with the value of 0 means that a given species doesn't exist in the given island whilst the value of 1 means that the species do exist in the given island.

The calculation of the co-occurrence of two species sp₁, sp₂ in the given set of islands is done as follows:

${\displaystyle C_{ij}=(r_{i}-S_{ij})(r_{j}-S_{ij})}$

C_ij - C-score for the two species sp₁, sp₂ in the given set of islands

S_ij - The number of co-occurrences of sp₁, sp₂

r_i - Number of islands in which sp₁ has 1

r_j - Number of islands in which sp₂ has 1

The checkerboard score (c-score) for the colonisation pattern is then calculated as the mean number of checkerboard units per species-pair in the community:

For M species, there are P = M(M-1)/2 species-pairs, so C-score is calculated:

${\displaystyle C=\sum _{j=0}^{M}\sum _{i<j}C_{ij}/P}$

The C-score is sensitive to the proportion of islands that are occupied, thereby confounding comparisons between matrices or sets of species pairs within them. An extension of the C-score therefore standardizes by the number of islands each species-pair occupies using:^[3]

${\displaystyle C_{ij}=(r_{i}-S_{ij})(r_{j}-S_{ij})/(r_{i}+r_{j}-S_{ij})}$