[R-sig-eco] Comparing slopes of SACs - Davies test

[Canning-Clode, Joao]

Dear list,

I am comparing species accumulation curves (in marine fouling communities) from 12 different ecoregions and I could use some help. One way to do this is to compare the asymptote's location (when fiting a non-linear curve). This is done. However I came across with this Davies test which I believe tests for a change in the slope - an initial break (Package: segmented). First question: can I only use this test for linear fits? I first fitted a linear model as a first step to identify the breakpoint:

> SF_LM <- lm(SF ~ Samples, data=SAC.data) ### fit linear model as first step to id breakpoint - need davies test next
> summary(SF_LM)

Call:
lm(formula = SF ~ Samples, data = SAC.data)

Residuals:
    Min      1Q  Median      3Q     Max
-8.0124 -0.8671  0.4903  1.4117  1.7396

Coefficients:
            Estimate Std. Error t value Pr(>|t|)
(Intercept) 13.19690    0.54598   24.17  < 2e-16 ***
Samples      0.21550    0.01863   11.56 1.76e-15 ***
---

Then I tried the test but I got this message:

> SF_LM_rr <- davies.test(SF_LM, "Samples")  ### find break point initial
Error in formula.default(eval(parse(text = x)[[1L]])) : invalid formula

Any help?
Also, do you recommend any other way (test) to compare these curves?

Thanks in advance

João

João Canning Clode, PhD
Postdoctoral Fellow
Marine Invasions Research Lab
Smithsonian Environmental Research Center
647 Contees Wharf Road
Edgewater, MD 21037

Email: canning-clodej at si.edu
Web: www.canning-clode.com
Tel: 443-482-2354