Table 2 The results of the logistic GLM estimation for clusters with high and low values (binary variables, hot and cold spots).

From: Subalpine woody vegetation in the Eastern Carpathians after release from agropastoral pressure

Variable

Estimate

Stand. Error

z

Marginal effect1

Wald test2

Constant

14.928

0.793

18.83

–

123.2

Altitude

 − 0.006

0.0008

 − 7.398

 − 0.0008***

175.0

Curvature

 − 0.336

0.048

 − 7.056

 − 0.045***

195.1

Ruggedness

3.895

0.105

37.21

0.526***

201.4

Distance

 − 1.290

0.049

 − 26.47

 − 0.174***

956.8

Land use 2

 − 0.375

0.279

 − 1.341

 − 0.044 n.s

45.6

Land use 3

 − 1.570

0.088

 − 17.73

 − 0.153***

263.2

Location 2

3.876

0.197

19.66

0.746***

23.6

Location 3

2.060

0.085

24.20

0.237***

956.8

Insolation

 − 5.746e−06

3.43E−07

 − 16.77

 − 7.756e−07***

137.8

Wind

 − 1.642

0.234

 − 7.026

 − 0.222***

175.0

Snow cover

0.0123

0.0008

14.47

0.002***

195.0

Snow distance

 − 0.009

0.001

 − 8.009

 − 0.001***

124.6

  1. Two variables were eliminated from the model: Exposure and Slope due to high collinearity with Ruggedness and Insolation. For the same reason, the categorical variables Land use 1 and Location 1 were automatically removed. For the description of the variables, see Table 1.
  2. 1***p < 0.001, n.s.—not significant.
  3. 2All values at p < 0.001.
Back to article page