Appendix C. Results of univariate ANOVA analyses of Katharina abundance under experimental devices, with effect sizes.
df SS MS F P Delta A) Summer 1998
Treatment 2 4.68 2.3 8.9 0.0009 0.56 Block 4 1.18 0.3 1.1 0.41 Error 8 2.11 0.3 contrasts Birds [-H +B -S (EC) vs. -H -B -S] 1 0.07 0.07 0.27 0.63 0.07 Shade [-H -B -S vs. -H -B +S] 1 3.96 3.96 15.0 0.005 0.51 B) November 1998
Treatment 2 1.75 0.9 2.80 0.12 0.34 Block 4 1.93 0.5 1.55 0.28 Error 8 2.50 0.3 contrasts Birds [-H +B -S (EC) vs. -H -B -S] 1 0.03 0.03 0.09 0.77 0.04 Shade [-H -B -S vs. -H -B +S] 1 1.11 1.11 3.54 0.10 0.27 C) Summer 1999
Treatment 2 3.16 1.6 3.1 0.10 0.46 Block 4 3.96 1.0 1.9 0.20 Error 8 4.14 0.5 contrasts Birds [-H +B -S (EC) vs. -H -B -S] 1 0.15 0.15 0.28 0.61 0.10 Shade [-H -B -S vs. -H -B +S] 1 1.73 1.73 3.34 0.11 0.34 D) November 1999
Treatment 2 0.78 0.4 2.08 0.19 0.23 Block 4 1.49 0.4 1.97 0.19 Error 8 1.51 0.2 contrasts Birds [-H +B -S (EC) vs. -H -B -S] 1 0.76 0.76 4.03 0.08 0.23 Shade [-H -B -S vs. -H -B +S] 1 0.09 0.09 0.49 0.50 0.08 E) Summer 2000
Treatment 2 8.70 4.4 25.8 0.0003 0.76 Block 4 2.83 0.7 4.2 0.04 Error 8 1.35 0.2 contrasts Birds [-H +B -S (EC) vs. -H -B -S] 1 0.04 0.04 0.3 0.63 0.05 Shade [-H -B -S vs. -H -B +S] 1 5.97 5.97 35.4 0.0003 0.63 Notes: Data are Katharina abundances under experimental devices within Bird Exclusion, Exclusion Control, and Artificial Shade treatments. Analyses are two-factor ANOVAs on square-root transformed means from summer (May, June, July, August) and fall (November) for 1998, 1999, and 2000. P values which are relevant to tested hypotheses and significant at the alpha = 0.01 level (adjusted for multiple comparisons) are in boldface type. Post hoc power analysis statistics were calculated using JMP 4.04. Delta = raw effect size. Specific treatment comparisons are provided for November analyses to illustrate power statistics despite the fact that the overall treatment effect was not significant (and therefore the alpha level for contrasts is actually greater than 0.05).