Table 2. Results of a logistic regression model, estimating how the probability of finding O. europaea depends on ontogenetic stage (adults [reference level], sapling, seedling) and environmental conditions and human disturbance (logging intensity) in Hugumburda dry Afromontane forest in northern Ethiopia.

	Estimate	SE	z	p
Intercept (Adult)	-0.136	0.07	-2.1	0.037
Sapling	-0.139	0.12	-1.0	0.317
Seedling	0.275	0.45	0.6	0.542
Woody species richness	0.269	0.08	3.2	0.001
Soil depth	-0.051	0.09	-0.6	0.554
Elevation	0.213	0.07	3.1	0.002
Slope	0.217	0.07	3.2	0.001
Logging intensity	0.253	0.08	3.3	0.001
Seedling × Woody species richness	-0.958	0.61	-1.6	0.114
Sapling × Woody species richness	0.467	0.17	2.8	0.006
Seedling × Soil depth	1.207	0.57	2.1	0.035
Sapling × Soil depth	-0.646	0.17	-3.8	<0.001
Seedling × Slope	0.580	0.45	1.3	0.201
Sapling × Slope	-0.324	0.13	-2.4	0.014
Seedling × Logging intensity	0.321	0.56	0.6	0.566
Sapling × Logging intensity	-0.687	0.19	-3.6	<0.001

The table shows parameter estimates and associated standard errors for a binary generalized linear model, where the response had two possible outcomes Y (empirical observation) = 1 and Y (simulated observation) = 0. Simulated observations were generated by random relocations of the empirical observations of O. europaea in 70 sample plots. Figures in bold are statistically significant at P < 0.05.