Skip to main content

Table 1 Non-equalized and group-equalized versions of the correlation indices (r) and the indicator value indices (IndVal)

From: The assignment of relevés to pre-existing vegetation units: a comparison of approaches using species fidelity

Indices

Non-equalized

Group-equalized

Correlation (r)

\(r_{\phi } = \frac{{N \times n_{p} - n \times N_{p} }}{{\sqrt {\left( {N \times n - n^{2} } \right) \times \left( {N \times N - N_{p}^{2} } \right)} }}\)

\(r_{\phi }^{g} = \frac{{N \times n_{p}^{g} - n^{g} \times N_{p}^{g} }}{{\sqrt {\left( {N \times n^{g} - n^{g2} } \right) \times \left( {N \times N_{p}^{g} - N_{p}^{g2} } \right)} }}\)

\(r_{ind} = \frac{{N \times a_{p} - a \times N_{p} }}{{\sqrt {\left( {N \times c \times a - a^{2} } \right) \times \left( {N \times N_{p} - N_{p}^{2} } \right)} }}\)

\(r_{ind}^{g} = \frac{{N \times a_{p}^{g} - a^{g} \times N_{p}^{g} }}{{\sqrt {\left( {N \times c \times a^{g} - a^{g2} } \right) \times \left( {N \times N_{p}^{g} - N_{p}^{g2} } \right)} }}\)

Indicator value (IndVal)

\(IndVal_{pa} = A_{pa} \times B_{pa} = \frac{{n_{p} }}{n} \times \frac{{n_{p} }}{{N_{p} }}\)

\(IndVal_{pa}^{g} = A_{pa}^{g} \times B_{pa} = \frac{{{{n_{p} } \mathord{\left/ {\vphantom {{n_{p} } {N_{p} }}} \right. \kern-\nulldelimiterspace} {N_{p} }}}}{{\sum\limits_{k = 1}^{k} {{{n_{k} } \mathord{\left/ {\vphantom {{n_{k} } {N_{k} }}} \right. \kern-\nulldelimiterspace} {N_{k} }}} }} \times \frac{{n_{p} }}{{N_{p} }}\)

\(IndVal_{ind} = A_{ind} \times B_{pa} = \frac{{a_{p} }}{a} \times \frac{{n_{p} }}{{N_{p} }}\)

\(IndVal_{ind}^{g} = A_{ind}^{g} \times B_{pa} = \frac{{{{a_{p} } \mathord{\left/ {\vphantom {{a_{p} } {N_{p} }}} \right. \kern-\nulldelimiterspace} {N_{p} }}}}{{\sum\limits_{k = 1}^{k} {{{a_{k} } \mathord{\left/ {\vphantom {{a_{k} } {N_{k} }}} \right. \kern-\nulldelimiterspace} {N_{k} }}} }} \times \frac{{n_{p} }}{{N_{p} }}\)

  1. Notes: rØ and rind are classified as correlation indices which the first one, phi coefficient, is calculated based on presence–absence data and the second one, correlation index for Individual-based, is derived from abundance data. IndValind is indicator value indices for species abundance data, and IndValpa is indicator value indices for presence–absence data. We follow mathematical formulae of the indices used in De Cáceres and Legendre (2009)
  2. N total number of relevés, Np number of relevés belonging to the target group, n number of occurrences of the species among all relevés, np number of occurrences of the species within the target group, ap sum of the abundance values of the species within the target group, a sum of the abundance values of the species over all relevés, c constant representing the total number of individuals or the total abundance per relevé. symbols are used in the Group-equalized indicator value indices, K number of groups, Nk number of relevés belonging to the kth group, nk number of occurrences of the species in the kth group, ak sum of the abundance values of the species in the kth group