2.4 Data analysis
Three-way analyses of variances (ANOVAs) were employed to test the main
and interactive effects of year, N addition, and P addition upon six
response variables: seed production, seed number per inflorescence,
inflorescence number, tiller number, density, and maximum height ofS . krylovii . Significant difference in means for seed
production, seed number per inflorescence, inflorescence number, tiller
number, density, and maximum height of S . krylovii among
the three levels of N or P addition were compared by Duncan’s multiple
range test. All the data were log-transformed to meet normality
assumptions for ANOVAs before these analyses, which were carried out
using SAS software (Proc Mixed, SAS 8.1; SAS Institute Inc., Cary, NC,
USA). In addition, confirmatory analyses based on structural equation
modeling (SEM) were conducted to quantify the direct and indirect
impacts of N and P addition upon seed production. This SEM analysis was
carried out in AMOS 21.0 (IBM, SPSS, Armonk, NY, USA).