. 2018 Nov 13;6:e5928. doi: 10.7717/peerj.5928

Table 2. The goodness of fit (R², please see Code 1 in Supplemental Files) of the fractal model (Su, 2016) on each of 19,833 samples are shown in the following table (the range in the “Average” column is standard error).

R² varies between 0 and 1, with larger numbers indicating better fits.

R²	Maximum	Minimum	Median	Average	Sample numbers
diatom	1.000	0.460	0.917	0.901 ± 0.001	3,224
fish	0.997	0.397	0.841	0.829 ± 0.004	761
bbs	0.982	0.555	0.792	0.791 ± 0.001	2,769
cbc	0.990	0.474	0.787	0.785 ± 0.002	1,999
fia	0.989	0.089	0.880	0.860 ± 0.001	10,355
gentry	0.969	0.640	0.911	0.892 ± 0.004	222
mcdb	0.981	0.466	0.869	0.838 ± 0.011	103
nabc	0.986	0.639	0.884	0.872 ± 0.004	400
Total	1.000	0.089	0.867	0.849 ± 0.001	19,833

Table 2. The goodness of fit (R2, please see Code 1 in Supplemental Files) of the fractal model (Su, 2016) on each of 19,833 samples are shown in the following table (the range in the “Average” column is standard error).