. 2015 Apr 20;10(4):e0123758. doi: 10.1371/journal.pone.0123758

Table 1. Weighted means of intercept (b _•0) and slope (b _•1) from the study-unit regressions described in S2 Table, for each of five types of interaction type and measured impact.

Interaction	Impacted	N	Intercept ^a	Slope ^a	SE intercept	SE slope	CI025 ^c	CI975
disease	plant	15	0.9694	-0.3906	0.0231	0.0104	-0.4110	-0.3701
herbivory	plant	12	0.9792	-0.4072	0.0498	0.0265	-0.4593	-0.3552
disease	enemy development	1	0.9965	-0.4042	NA ^b	NA	NA	NA
herbivory	enemy development	9	0.9818	-0.0766	0.0201	0.0113	-0.0988	-0.0545
herbivory	enemy behavior	11	0.9961	-0.2520	0.0297	0.0162	-0.2836	-0.2203

^aCoefficients are for the model RI = b _•0 + b _•1log₁₀(PD + 1), where PD is the phylogenetic distance in Ma from the most strongly affected host.

^bStandard errors for the weighted means are not available for disease—enemy development category because there was only one study unit (N), but were significant in the original study.

^cThe 95% confidence intervals are presented for the weighted mean slopes.

Table 1. Weighted means of intercept (b •0) and slope (b •1) from the study-unit regressions described in S2 Table, for each of five types of interaction type and measured impact.

Table 1. Weighted means of intercept (b _•0) and slope (b _•1) from the study-unit regressions described in S2 Table, for each of five types of interaction type and measured impact.