Table 4 Results of maximum likelihood population effects (MLPE) model selection after accounting for non-independence of pairwise distance observations

Model	Regression slope Est.±95% Confidence Intervals				Marginal R²
	Distance	Elevation	Land cover	Land-elevation
(a)
4	—	—	—	0.83±0.04	0.879
9	0.01±0.01	—	—	0.83±0.04	0.878
7	—	2.05±5.42	0.83±0.04	—	0.875
6	0.01±0.01	—	0.83±0.04	—	0.875
3	—	—	0.89±0.04	—	0.875
8	0.03±0.02	−7.38±10.72	0.81±0.04	—	0.874
5	0.23±0.11	94.38±45.94	—	—	0.158
1	0.05±0.06	—	—	—	0.022
2	—	−12.00±27.41	—	—	0.008
(b)
9	0.08±0.05	—	—	0.59±0.14	0.711
6	0.08±0.05	—	0.59±0.14	—	0.710
7	—	24.19±20.31	0.65±0.14	—	0.709
8	0.12±0.10	−17.46±42.29	0.56±0.16	—	0.694
4	—	—	—	0.06±0.37	0.667
3	—	—	0.64±0.15	—	0.660
5	0.27±0.11	−80.24±48.16	—	—	0.173
1	0.11±0.07	—	—	—	0.129
2	—	16.96±29.46	—	—	0.022

For all models, pairwise (a) F_ST and (b) D_EST values were the response variable. Models are ranked based on marginal R² value (high to low) with the best fitting model having the highest R² value. For each explanatory variable included in the model, values (× 10^–4) are presented as regression slopes±95% confidence interval. Values in bold indicate significant parameter estimates (that is, 95% confidence intervals that do not overlap zero). Explanatory variables not included in the model are indicated by ‘—’.

Quick links

Search