Research Article

Structural Equation Modeling: Theory and Applications in Forest Management

Table 5

Estimated total effects (i.e., direct path and all possible indirect paths) on the endogenous latent variables in the overstory-understory SR model, along with corresponding standard errors and values.

Path modelTotal effectStandard error valued

Late-seral herb cover (LSHERB) =
  Overstory cover (TREE)−0.2440.0690.000
  Tree size (QMD)−0.0280.0230.235
  Tree density (ln(TPH) × 10)−0.3610.2160.095
  Stand age (AGE)0.1600.0430.000
  Aspect (ASPECT × 10)0.7250.1870.000
  Fine litter (FLITTER)0.2730.1810.132
  Coarse litter (CLITTER)0.1490.2160.490
  Understory cover (UNDER)0.1960.0460.000

Fine litter (FLITTER) =
  Overstory cover (TREE)0.1350.0310.000
  Tree size (QMD)−0.1020.0380.007
  Tree density (ln(TPH) × 10)−0.4740.1200.000
  Stand age (AGE)−0.0010.0010.327
  Aspect (ASPECT × 10)0.1110.0530.038
  Coarse litter (CLITTER)−0.0200.0200.296
  Understory cover (UNDER)0.0720.0120.000

Coarse litter (CLITTER) =
  Tree density (ln(TPH) × 10)0.1700.0550.002
  Stand age (AGE)0.0300.0110.007

Understory cover (UNDER) =
  Overstory cover (TREE)−0.5080.1810.005
  Tree density (ln(TPH) × 10)−1.0370.4000.010
  Stand age (AGE)−0.0090.0090.337
  Aspect (ASPECT × 10)1.5430.4180.000
  Coarse litter (CLITTER)−0.2840.2810.313

dStatistical tests were based on a t-distribution with 1180 degree of freedom as the effects of the nesting structure on standard errors have been adjusted.