Structural Equation Modeling: Theory and Applications in Forest Management
Table 4
Estimated combined indirect effects (i.e., all possible indirect paths) on the endogenous latent variables in the overstory-understory SR model, along with corresponding standard errors and values.
Path model
Indirect effect
Standard error
valuec
Late-seral herb cover (LSHERB) =
Overstory cover (TREE)
−0.053
0.057
0.353
Tree size (QMD)
−0.028
0.023
0.235
Tree density (ln(TPH) × 10)
−0.277
0.091
0.003
Stand age (AGE)
0.005
0.006
0.478
Aspect (ASPECT × 10)
0.302
0.126
0.017
Coarse litter (CLITTER)
−0.056
0.053
0.295
Understory cover (UNDER)
0.020
0.014
0.165
Fine litter (FLITTER) =
Overstory cover (TREE)
−0.036
0.016
0.022
Tree density (ln(TPH) × 10)
−0.074
0.028
0.007
Stand age (AGE)
−0.001
0.001
0.327
Aspect (ASPECT × 10)
0.111
0.037
0.003
Coarse litter (CLITTER)
−0.020
0.020
0.296
Understory cover (UNDER) =
Tree density (ln(TPH) × 10)
−0.048
0.047
0.308
Stand age (AGE)
−0.009
0.009
0.338
cStatistical tests were based on a -distribution with 1180 degree of freedom, with adjustments for the effects of the nested sampling structure on standard errors.