Ecological Archives E085-108-A2

Pierre Legendre, Mark R. T. Dale, Marie -Josée Fortin, Philippe Casgrain, and Jessica Gurevitch. 2004. Effects of spatial structures on the results of field experiments. Ecology 85:3202–3214.

Appendix B. Type I error in simulations involving n = 144 (Fig. B1) and n = 36 (Fig. B2) experimental units.

   FIG. B1. Rate of Type I error in simulations involving n = 144 experimental units after 1000 simulations. The rows of diagrams contain the results for the 6 experimental designs, numbered 1 to 6. Left-hand column: results for simple ANOVA. Middle and right-hand column: results for ANOVA taking blocking into account, statistics F1 and F2. The 4 types of environmental surfaces are represented by symbols. Abscissa: ranges of values of spatial autocorrelation (SA) in the environmental (Envir.) and response variables (Resp.). Asterisks: rejection rates whose confidence intervals do not include the significance level (alpha = 0.05).

Original image available for download as "Fig_B1.gif". Reduce figure to 25% in Page Setup before printing.

 


   FIG. B2. Rate of type I error in simulations involving n = 36 experimental units after 1000 simulations. The rows of diagrams contain the results for the 6 experimental designs, numbered 1 to 6. Left-hand column: results for simple ANOVA. Middle and right-hand column: results for ANOVA taking blocking into account, statistics F1 and F2. The 4 types of environmental surfaces are represented by symbols. Abscissa: ranges of values of spatial autocorrelation (SA) in the environmental (Envir.) and response variables (Resp.). Asterisks in panels 1–4: rejection rates whose confidence intervals do not include the significance level (alpha = 0.05).

Original image available for download as "Fig_B2.gif". Reduce figure to 25% in Page Setup before printing.

 



[Back toE085-108]